Breast microcalcifications are key radiological indicators of malignancy, and their assessment in denser breast tissues is challenging. We propose a Wavelet-based Attention Deep Network (WADN) that isolates high-frequency components linked to microcalcification clusters using denoising wavelet transforms and integrates them into a deep attention architecture for interpretable, localization-assisted classification. The system is deployed in an edge-computing framework enabled by Jetson Orin Nano to support real-time analysis and remote radiologist access for the Medical Internet of Things (MIoT). For the private PINUM and public DDSM datasets, WADN outperforms standard convolutional neural networks (CNNs) and classical machine learning baselines, achieving accuracies of 0.95 and 0.93, respectively, with consistent gains in sensitivity, specificity, precision, F1-score, and area under the curve (AUC). On the IoT edge device, the method achieves accuracies of 0.91 for PINUM and 0.93 for DDSM, demonstrating the feasibility of ondevice screening. These results indicate that combining wavelet transforms with attention mechanisms and IoT edge deployment improves diagnostic accuracy while enabling practical real-time breast-cancer screening workflows.

Social recommendation leverages social relations to complement user–item interactions, alleviating data sparsity in recommendation systems. However, existing approaches still suffer from two critical issues: social graphs inevitably contain noisy or redundant edges, which can hinder user representation learning, and the results of recommendations are biased toward popular items, failing to capture user preferences for long-tail items. To address these challenges, we propose a novel model named denoised and bias-aware framework for social recommendation (DBA-SR). DBA-SR incorporates two key innovations. First, a structure-aware social graph pruning module adaptively removes weak or redundant edges to construct a denoised and structurally informative social graph. Second, a tail- and popularity-aware preference modeling mechanism jointly enhances long-tail preference learning and suppresses popularity bias through dedicated regularization strategies. The overall model is trained under a unified dual graph reconstruction and variational regularization objective. Extensive experiments on three real-world benchmark datasets demonstrate that DBA-SR consistently outperforms baselines in top-$N$recommendation tasks. Further ablation studies confirm the complementary effects of structure-aware pruning and preference specialization. This work provides new insights into the role of structural denoising in social graphs and demonstrates an effective solution to mitigating popularity bias.

Architectural Distortion (AD) is a common abnormality in digital mammograms, alongside masses and microcalcifications. Detecting AD in dense breast tissue is particularly challenging due to its heterogeneous asymmetries and subtle presentation. Factors such as location, size, shape, texture, and variability in patterns contribute to reduced sensitivity. To address these challenges, we propose a novel feature fusion-based Vision Transformer (ViT) attention network, combined with VGG-16, to improve accuracy and efficiency in AD detection. Our approach mitigates issues related to texture fixation, background boundaries, and deep neural network limitations, enhancing the robustness of AD classification in mammograms. Experimental results demonstrate that the proposed model achieves state-of-the-art performance, outperforming eight existing deep learning models. On the PINUM dataset, it attains 0.97 sensitivity, 0.92 F1-score, 0.93 precision, 0.94 specificity, and 0.96 accuracy. On the DDSM dataset, it records 0.93 sensitivity, 0.91 F1-score, 0.94 precision, 0.92 specificity, and 0.95 accuracy. These results highlight the potential of our method for computer-aided breast cancer diagnosis, particularly in low-resource settings where access to high-end imaging technology is limited. By enabling more accurate and timely AD detection, our approach could significantly improve breast cancer screening and early intervention worldwide.

Graph-based Retrieval-Augmented Generation (GraphRAG) is a prominent technique for mitigating hallucinations in Large Language Models (LLMs), yet it faces critical limitations in high-stakes anesthesiology question answering. At the retrieval stage, pervasive synonymy and knowledge fragmentation cause clinical queries to resolve to incorrect synonym clusters, while the highly specialized and domain-specific nature of clinical language widens the semantic gap, making an effective balance between recall and precision difficult to achieve. At the reasoning stage, the high structural heterogeneity of anesthesiology knowledge graphs, combined with the strong context-dependence of clinical semantics, renders existing multi-hop reasoning susceptible to clinical logic drift along weak associations, thereby introducing unreliable reasoning paths. Finally, at the fusion stage, conflicts between retrieved evidence and the model's parametric knowledge can easily lead to highly confident yet clinically unsafe hallucinations. To address these challenges, we propose the Description-augmented, Uncertainty-aware, and Adaptive Layered Knowledge Framework (DUAL-Know). For retrieval, query augmentation and description-augmented semantic recall bridge terminology gaps and comprehensively retrieve relevant knowledge. For reasoning, a Dynamically Gated Heterogeneous Multi-Head Attention (DGHMA) mechanism accommodates graph heterogeneity and dynamically injects query intent into each reasoning layer, followed by an uncertainty-aware path ranking module that filters noisy chains. For fusion, a multi-metric scoring strategy jointly evaluates model confidence, retrieval consistency, and semantic coherence to safely arbitrate knowledge conflicts. Experiments demonstrate that DUAL-Know consistently outperforms strong baselines, generating more accurate, reliable, and verifiable answers for complex clinical question-answering tasks.

The diagnosis of high myopia using fundus images is essential for visual health. Existing deep learning-based methods rely on large-scale labeled data, but annotated data for high myopia fundus images remains scarce. To address this issue, we mimic the ability of ophthalmologists to diagnose a new disease with only a small number of samples. In this regard, we propose a meta-learning network guided by domain knowledge from fundus images for diagnosing high myopia. The model consists of three modules: First, the image reconstruction module builds an autoencoder (comprising an encoder and a decoder) that takes the input raw image and outputs the reconstructed image. Next, the fundus image domain knowledge learning module constructs a Siamese network to learn the similarity between the original and reconstructed fundus images. This similarity is used as a loss function to guide the encoder in effectively learning the domain features of fundus images. Finally, in the domain-knowledge-guided meta-learning module, the encoder’s initialization parameters (obtained from the first two modules) are further optimized using the OCMAMAL architecture, resulting in more optimal encoder parameters. This optimization helps achieve superior recognition performance with only a small amount of data for new tasks. Using a clinically real high myopia fundus image dataset, our method achieved F1 scores of 83.4%, 87.9%, and 89.2% under 5-shot, 10-shot, and 20-shot conditions, respectively, demonstrating the effectiveness of the proposed method.

Brain tumor localization and segmentation from MRI is a challenging task in the field of Medical AI. With recent advancements, various techniques have been developed to assist medical professionals in detecting brain tumors using Artificial Intelligence. Although machine learning algorithms have shown efficiency in tumor segmentation, they often lack interpretability, making it difficult to trust and validate their predictions. In this paper, we developed an interpre table UNet model for brain tumor segmentation, incorporating Gradient-weighted Class Activation Mapping (Grad-CAM) and Shapley Additive Explanations (SHAP) to enhance model transparency. The BraTS2020 benchmark dataset was used for training and evaluation. The model demonstrated strong performance, achieving an accuracy of 99.35%. Grad-CAM was applied to visualize the regions of interest identified by the model, enhancing interpretability. Additionally, the SHAP library was utilized to explain the predictions of multiple machine learning models (Random Forest, KNN, SVC, MLP) employed for estimating patient survival days, further contributing to the model’s transparency and clinical utility.

Cushing’s syndrome is caused by excessive glucocor-ticoid secretion and often presents with facial features such as moon facies and plethora, making facial images valuable for diagnosis. Recent studies have used pre-trained CNNs for automated diagnosis using frontal facial images. However, CNNs focus on local features and may miss global facial characteristics typical of Cushing’s syndrome. Transformer-based visual models, through self-attention mechanisms, are better suited for capturing global features. The foundational model DINOv2, also based on the vision Transformer architecture, has recently attracted attention. In this study, we compared the performance of various pre-trained models, including CNNs, Transformer-based models, and foundational models like DINOv2, in a transfer learning setting. We also investigated biological sex bias and the effect of freezing mechanisms on DINOv2. Our results show that Transformer-based models, especially ViT, and DINOv2 outperformed CNNs, with ViT achieving the highest F1 score of 85.74%. All models showed higher accuracy for female samples, likely due to biological sex imbalance. Freezing mechanisms notably improved DINOv2 performance. In conclusion, Transformer-based models and DINOv2 show strong potential for classifying Cushing’s syndrome using facial images.The source code is publicly available at: https://github.com/songchangwei/Cushing-Disease-Diagnosis.

Delirium is an acute and reversible neuropsychiatric syndrome, with its diagnosis relying on clinicians’ dynamic assessments of patients. This process requires the integration of multidimensional clinical information, for which electronic medical records (EMRs) serve as a crucial source. However, Chinese EMRs are often lengthy, unevenly structured, and exhibit considerable variability in symptom descriptions, leading to key diagnostic information being obscured by redundant narratives. These characteristics pose significant challenges for traditional methods to consistently extract relevant diagnostic features. This paper proposes a deep learning framework, guided by medical knowledge, to improve the classification performance and interpretability of delirium in Chinese long-form EMRs. The framework employs a region-guided chunking mechanism to segment records into semantically coherent units and extract key features. It combines differentiated data augmentation and contrastive learning methods to strengthen semantic discriminability, and utilizes adaptive feature fusion to achieve cross-chunk collaborative decision-making. Experimental results demonstrate that this method effectively enhances the accuracy and reliability of delirium diagnosis in Chinese medical texts, showing strong clinical applicability.

Computerized tomography (CT) provides non-invasive visualization of internal structural information without losing any detail. It has proven to be very useful in protecting cultural relics. However, metal cultural relics are frequently accompanied by destructive metal artifacts in x-ray CT images, making it impossible for traditional methods to obtain detailed information from the cultural relics. In recent years, deep generative-based models have demonstrated great promise for solving such problems. However, due to the complicated structure and diverse materials of cultural relics, it is difficult to accurately restore the highly heterogeneous details of cultural relics in practical applications. As a result, the primary focus of this study is on the removal of metal items from cultural relics using deep generative models and achieves effective restoration of highly heterogeneous details by combining mask-guided strategies. Specifically, we first collaborated with the Palace Museum to build a cultural relic CT dataset and specifically divided artifacts into two categories: sharp edge smoothing and edge distortion according to their complexity. Second, many deep generative networks include CycleGAN, CSGAN, MUNIT, DeblurGAN, and DRIT were trained. Additionally, segmentation masks were blended to create artifact-free images. Finally, the performance was enhanced via dynamic weight adjustment. The effectiveness has been qualitatively and quantitatively validated on the cultural relic CT dataset. PSNR and SSIM metrics confirm the model’s ability to restore fine details, providing reliable support for future cultural relic protection.

Mental health is a significant global public health issue, and psychological support hotlines play a crucial role in providing mental health assistance and identifying suicide risks at an early stage. However, the emotional expressions conveyed during these calls remain underexplored in current research. This study introduces a novel method that combines pitch acoustic features with deep learning-based features to analyze and understand emotions expressed during hotline interactions. Using data from China's largest psychological support hotline, which includes 105 subjects, our method achieved an F1-score of 79.13% for negative binary emotion classification. Additionally, the proposed approach was validated on an open dataset for multi-class emotion classification, where it demonstrated better performance compared to the state-of-the-art methods. To explore its clinical relevance, we applied the model to analysis the frequency of negative emotions and the rate of emotional change in the conversation, comparing 46 subjects with suicidal behavior to those without. While the suicidal group exhibited more frequent emotional changes than the non-suicidal group, the difference was not statistically significant. Importantly, our findings suggest that emotional fluctuation intensity and frequency could serve as novel features for psychological assessment scales and suicide risk prediction. The proposed method provides valuable insights into emotional dynamics and has the potential to advance early intervention and improve suicide prevention strategies through integration with clinical tools and assessments. The source code is publicly available at: https://github.com/Sco-field/Speechemotionrecognition/tree/main.

Diabetic retinopathy (DR) is the leading cause of blindness in diabetic patients, which necessitates precise grading of retinal lesions for early diagnosis. Existing DR grading methods typically employ image enhancement techniques to improve the quality of fundus images. However, due to variations in imaging devices and differences in the proficiency of medical practitioners, the quality of images often exhibits significant heterogeneity. Uniform enhancement across all fundus images may inadvertently amplify noise artifacts, particularly in high-quality images. Moreover, since diabetic lesions in fundus images are often small, reliance solely on global image features makes it difficult to fully capture fine-grained lesion features. To address these challenges, this paper proposes a dual-stream deep learning model that integrates quality-aware dynamic enhancement and a multi-instance multi-scale vision transformer. First, An image quality assessment-based selective enhancement strategy was implemented, wherein only low-quality fundus images underwent enhancement processing. Then, a dual-branch processing architecture is designed to differentially handle enhanced and non-enhanced images. Experimental results on real-world datasets demonstrate the effectiveness of the proposed method.

Neurological disorders are the leading cause of global disease burden. Early diagnosis and treatment are crucial for enhancing patients’ quality of life and improving their prognosis. With the rapid development of artificial intelligence, especially in speech processing technology, the diagnosis of neurological diseases has made revolutionary breakthroughs. Speech processing technology can detect subtle pronunciation variations in patients’ pronunciation, enabling doctors to identify neurological disorders more objectively and accurately. In this paper, we focus on two specific disorders (Parkinson’s disease, Alzheimer’s disease), conducting an in-depth analysis and comprehensive summary of the existing research on speech-based diagnosis in neurological diseases.This article starts by introducing commonly used public speech datasets and AI processing frameworks. Then, from the perspective of model construction, this article elucidates advanced endeavors in the diagnosis of Parkinson’s disease and Alzheimer’s disease. The related works are divided into three categories: machine learning, deep learning, and large models. Furthermore, this article elaborates on the existing problems in current research and the future development trends, aiming to offer novel insights into the in-depth application of speech processing technology in neurological diseases.

In the medical treatment of pneumonia, most patients undergo multiple CT scans at different time points, generating serialised CT data. However, traditional segmentation methods based on convolutional networks are difficult to capture inter-slice dependencies, which can result in missing sequence information. Additionally, Transformer-based segmentation methods tend to overemphasize temporal information, leading to the loss of local features in a single image. To address these issues, we propose a Serialised Pulmonary Lesion Segmentation with Spatiotemporal Transformer (SPLS-STT). Specifically, we firstly capture local features of lesion regions and global features from the serialised lung CT through a self-attention mechanism, which generates a spatial feature maps. Secondly, the obtained spatial feature maps at multiple time points are arranged in chronological order, and the dependencies between the pneumonia lesions at different time points are established by the cross-window temporal attention mechanism. Finally, we combine the features in the temporal and spatial dimensions through the cross-attention mechanism to compensate for the loss of local features during feature extraction in the temporal dimension. Experimental results show that our spatiotemporal Transformer model outperforms the traditional single-dimensional processing methods in terms of accuracy metrics. Furthermore, our model achieves a Dice coefficient of 37.9% on the lung CT sequence dataset, which is an improvement of 9.5% compared to the Vision Transformer.

Wei Zhai, Nan Bai, Qing Zhao, Jianqiang Li, Fan Wang, Hongzhi Qi, Meng Jiang, Xiaoqin Wang, Bing Xiang Yang, Guanghui Fu. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Deep learning performs feature extraction through a series of data transformations. Convolutional neural networks (CNNs) are among the most representative methods in deep learning. CNNs enable complex data transformations through convolution operations to extract high-level features. The kernel-based autoencoder (KAE) uses kernelized principal component analysis to perform both linear and nonlinear data transformations. Consequently, KAE provides better interpretability than CNNs in feature extraction. This paper presents a comprehensive study of KAE, including both single-layer and multi-layer configurations. Two distinct multi-layer autoencoder architectures based on the kernel method are proposed. The proposed method is evaluated using handwritten digit images. Experimental results demonstrate the proposed method’s effectiveness in feature extraction. In terms of the structural similarity index measure, the multi-layer kernel-based autoencoder achieves results comparable to those of the convolutional autoencoder. With fewer training samples, the multi-layer kernel-based autoencoder outperforms the convolutional autoencoder.

Accurate prediction of intraoperative hazardous events and generation of effective intervention plans are critical to surgical safety, but face multiple challenges of real-time, accuracy, and interpretability. Large-scale language models have potential, but their high cost and potential ‘illusion’ problems limit their application in real-time clinical environments. Traditional multitask learning models are efficient but knowledge-constrained, making it difficult to capture complex reasoning processes. To bridge this gap, this paper proposes a multi-objective distillation knowledge enhancement model-KEMO, which innovatively adopts a multi-objective chain-of-thought distillation framework to not only mimic the prediction results of the instructor’s LLM, but also explicitly migrate its structured reasoning process to the lightweight student model, which improves the answerability of the model by synergistically optimising the three objectives of event prediction, reasoning alignment and scenario generation. Interpretability. Meanwhile, combined with the Knowledge Graph-based Retrieval Augmented Generation mechanism, validated medical knowledge is dynamically injected to enhance the accuracy and reliability of decision-making and reduce model illusion. The experimental results show that the KEMO model significantly outperforms traditional models of the same magnitude in intraoperative hazardous event prediction and prognostic proposal generation, and achieves a performance comparable to that of a large faculty model.The KEMO model effectively bridges the gap between the large language model and the actual clinical application, and facilitates the transformation of the large model knowledge to the actual clinical deployment.

In evolutionary computation, the study subject for algorithms capable of effectively resolving multi-objective optimization problems remains at the forefront of research. Most existing multi-objective evolutionary algorithms (MOEAs) often struggle with maintaining diversity and avoiding premature convergence when solving complex or high-dimensional optimization problems. This study proposes an innovative iteration of the Non-dominated Sorting Genetic Algorithm III (NSGA-III), which infuses chaotic dynamics alongside estimated convergence point strategies to enhance solution quality and diversity. Our research undertakes a comprehensive evaluation of this enhanced algorithm against a suite of benchmark problems. We compare it with its predecessor, Non-dominated Sorting Genetic Algorithm II (NSGA-II), and variations incorporating chaotic dynamics alongside estimated convergence point strategies. The performance analysis results indicate that NSGA-III, using chaotic dynamics and estimated convergence point strategies across most test functions, demonstrates superior performance over traditional and singly-enhanced MOEAs. The statistical analysis strongly suggests that the dual enhancements embedded in the novel algorithm contribute significantly to its optimization capabilities.

Feature sparse problem is commonly existing in patient similarity calculation task with clinical data, to track which, some approaches have been proposed to use Graph Neural Network (GNN) to model the complex structural information in patient Electronic Medical Records (EMRs). These GNN based approaches usually treat medical concepts (i.e., symptoms, diseases) as nodes to learn spatial features and adopt Recurrent Neural Network (RNN) to learn temporal sequence of these concepts. However, in many cases, several sequential concepts contained in EMR text are considered as occur simultaneously in the clinical diagnosis (i.e., some symptoms are detected simultaneously by once test), learning temporal sequence of these sequential concepts might cause noise for patient similarity calculation. Furthermore, the limited discriminative capability of concepts cannot provide sufficient indicative information for similarity learning. To this end, we propose a Knowledge-guided Event-relation Graph Learning Network (KEGLN) for patient similarity calculation. Specifically, after event extraction, we first construct element-relation graphs and use the first Graph Convolutional Network (GCN) and Graph Attention Network (GAT) layer to aggregate features from each event and its involved elements for reducing the noise produced by temporal sequence of concepts. Meanwhile, the entity description and attribute-value structure are extracted to supplement background knowledge of elements (concepts and trigger words). For the updated event nodes, we then design a event-relation graph and adopt the second GCN and GAT layer to aggregate information from events and their directly neighbors to extract spatial features of events at the current moment. Finally, the Bidirectional Long Short-Term Memory (BiLSTM) model is adopted to learn temporal dependency of event nodes to capture the dynamic change of disease progress. Through diverse datasets and extensive experiments, our KEGLN model outperforms all baselines for Chinese patient similarity calculation.

Non-invasive fundus images can be used to diagnose various fundus diseases, such as high myopia (HM). Existing deep learning-based research mainly relies on data to drive the model to learn key features. However, the data related to HM is limited (especially for young children), making it difficult for deep networks to accurately focus on key features. Hence, we propose a prior knowledge-guided deep learning network for pediatric HM detection. It comprises four modules: (1) Prior Feature-Based Channel Fusion: This module extracts key features (brightness, edges, texture) from fundus images using image processing methods to obtain corresponding single-channel slices. Through channel-level feature fusion, these slices are used to construct multiple sets of feature-enhanced datasets. (2) Global Fundus Feature Extraction: It uses residual blocks to build the backbone, and builds aU-shaped attention component based on the U -shaped network. This module extracts the global and context information of the original fundus image to obtain a global feature map. (3) Knowledge-Guided Attention Generation: The residual structure is employed to further extract the hidden features of the feature-enhanced data, thereby obtaining local key feature maps. (4) Pediatric HM Classification: By combining local key feature maps (obtained in module 3) with global feature maps (obtained in module 2) through spatial attention mechanism, the deep network is guided to complete the classification task of pediatric HM. Extensive experiments on real-world datasets demonstrate the effectiveness of our method (accuracy is 0.921, F1 score is 0.903).

Medical image analysis frequently encounters data scarcity challenges. Transfer learning has been effective in addressing this issue while conserving computational resources. The recent advent of foundational models like the DINOv2, which uses the vision transformer architecture, has opened new opportunities in the field and gathered significant interest. However, DINOv2's performance on clinical data still needs to be verified. In this paper, we performed a glioma grading task using three clinical modalities of brain MRI data. We compared the performance of various pre-trained deep learning models, including those based on ImageNet and DINOv2, in a transfer learning context. Our focus was on understanding the impact of the freezing mechanism on performance. We also validated our findings on three other types of public datasets: chest radiography, fundus radiography, and dermoscopy. Our findings indicate that in our clinical dataset, DINOv2's performance was not as strong as ImageNet-based pre-trained models, whereas in public datasets, DINOv2 generally outperformed other models, especially when using the frozen mechanism. Similar performance was observed with various sizes of DINOv2 models across different tasks. In summary, DINOv2 is viable for medical image classification tasks, particularly with data resembling natural images. However, its effectiveness may vary with data that significantly differs from natural images such as MRI. In addition, employing smaller versions of the model can be adequate for medical task, offering resource-saving benefits. Our codes are available at https://github.com/GuanghuiFU/medical_dino_eval.

Depression is a prominent public mental health issue affecting millions globally. With the growth of online communities, many individuals suffering from depression are increasingly seeking guidance and support from community-based Question-Answering (QA) systems. Previous methods, which recommend the best answer by calculating the semantic distance between a question and its potential answers, seldom incorporate psychological indicators as features. Answers from non-professionals might offer misguided information to those with depression. In our approach, we employ a multi-task learning method, taking into account both semantic information and psychological indicators for answer selection in depression community QA systems. To compute the semantic similarity between a given question and its potential answers, we gather surface-layer textual information. We then evaluate these candidate answers using psychological indicators from two dimensions: (1) employing five expert-defined comprehensive quality indicators to assess answer quality, and (2) using existing standard answers, provided by experts for various question categories, as external references to automatically review the candidate answers. Finally, by integrating the semantic similarity and psychological indicators, we make the final answer selection. Experimental results have shown that our model significantly enhances answer recommendations in the mental health domain.

Breast cancer has become the primary cancer endangering women's life and health. A large number of evidence based on medicine show that early screening can effectively reduce the mortality rate of breast cancer. With the development of computer vision technology, the computer-aided diagnosis system for breast cancer screening and detection has attracted extensive attention from all walks of life and breast lesion localization and diagnosis are the key steps. Therefore, this article starts from the perspective of model feature construction to conduct an in-depth analysis and a comprehensive summary of the existing research on breast image lesion localization and benign-malignant diagnosis. This article divides feature construction methods into three types based on novel perspectives: domain knowledge-driven, data-driven, and domain knowledge-driven fusion data-driven. On this basis, the article uses a systematic review method to classify, summarize, and compare the models for breast image lesion localization and diagnosis which greatly expands and deeply analyzes existing reviews. In addition, this article elaborates on the existing problems in current research work and discusses the future outlook of breast image lesion localization and benign-malignant diagnosis.

The inherent sparsity of electronic medical record (EMR) poses difficulties for the learning of patient similarity. Graph-based modeling methods can infer missing values by learning complex relations among medical facts, which are becoming the mainstream options for patient similarity analysis. However, existing graph-based solutions mainly focus on general patterns among patients and overlook local differences, potentially losing semantic information on patient similarity. Additionally, in real medical situations, entities related to symptoms or treatment types often involve multiple diseases, which provide erroneous signals in similarity assessment. Therefore, this paper proposes a novel deep learning method called Multi-view Hierarchical Learning Network (MHLN) for patient similarity measurement. This method extracts dependency information between entities from both local and global perspectives. Additionally, it assigns different importance levels to different types of medical entities, enabling the learning of dependency features between entity representations through local and global encoders to generate representative embeddings for similarity computation. Finally, we evaluate MHLN on real-world Chinese EMR data, and the results demonstrate the effectiveness of MHLN compared to related work.

At the end of 2019, the COVID-19 outbreak emerged abruptly. Chinese health authorities highlighted the role of CT scans, X-rays, and other computerized lung imaging in aiding COVID-19 diagnosis. This study aims to develop a computer-based system to assist healthcare professionals in diagnosing COVID-19 infections based on computerized imaging analysis. This approach aims to alleviate the workload of COVID-19 specialists, improving diagnostic and treatment efficiency and allowing specialists to focus on devising appropriate patient care plans promptly. The proposed method focuses on analyzing COVID-19 lesion characteristics within individual CT slices and their serial characteristics across CT sequences. This approach mirrors the diagnostic process of radiologists closely. To validate our model, we compiled a dataset from real medical diagnostic settings, minimizing the impact of lesion-like artifacts. We conducted a series of comparative and ablation experiments to evaluate the model's performance. Results indicate that our model outperforms the classic classification models and other commonly used models for COVID-19 diagnosis on our constructed dataset.

Multi-class segmentation of eye images plays a pivotal role in assessing patients with myasthenia gravis, and the measurement results rely heavily on the segmentation accuracy. However, there is still a problem with inaccurate boundary segmentation. Compared to heuristic-based network structure optimization, exploring effective loss function is an intuitive, simple, and interpretable way to address this issue. In this paper, we experimentally verify the effectiveness of boundary loss for multi-class segmentation of eye images and investigate its hybrid law with other segmentation losses. The application of the study significantly enhances the accuracy of myasthenia gravis scoring and holds promise for assisting in the evaluation of various other eye diseases.

Sign language video translation, which converts sign language information into textual expressions, play a vital role in breaking down the language communication barrier between deaf and healthy people. Existing translation methods are mainly focus on the single and pure background. However, the background in real-world environments is always complex, and these methods are difficult to achieve effective recognition results. To address this issue, we have exploratively constructed a real-world complex background sign language dataset (CBSL), containing sign language videos captured in various authentic environments (e.g., different backgrounds and lighting conditions). Based on this, we propose a progressive sign language translation model to effectively separate sign language users from the background and reduce environmental interference, thus significantly improving the generalization ability. Our proposed method significantly outperforms various comparative methods across all performance metrics on the CBSL dataset. Furthermore, on the publicly available Chinese Sign Language Continuous Recognition dataset(CSL), our method performs comparably to the current state-of-the-art (SOTA).

Suicide and suicidal behaviors remain significant challenges for public policy and healthcare. In response, psy-chological support hotlines have been established worldwide to provide immediate help to individuals in mental crises. The effectiveness of these hotlines largely depends on accurately identifying callers' emotional states, particularly underlying negative emotions indicative of increased suicide risk. However, the high demand for psychological interventions often results in a shortage of professional operators, highlighting the need for an effective speech emotion recognition model. This model would automatically detect and analyze callers' emotions, facil-itating integration into hotline services. Additionally, it would enable large-scale data analysis of psychological support hotline interactions to explore psychological phenomena and behaviors across populations. Our study utilizes data from the Beijing psychological support hotline, the largest suicide hotline in China. We analyzed speech data from 105 callers containing 20,630 segments and categorized them into 11 types of negative emotions. We developed a negative emotion recognition model and a fine-grained multi-label classification model using a large-scale pretrained model. Our experiments indicate that the negative emotion recognition model achieves a maximum F1-score of 76.96%. However, it shows limited efficacy in the fine-grained multi-label classification task, with the best model achieving only a 41.74% weighted F1-score. We conducted an error analysis for this task, discussed potential future improvements, and considered the clinical application possibilities of our study. All the codes are public available at: https://github.com/cz10914/psy_hotline_analysis.

We propose an extension of the chaotic evolution algorithm into the discrete domain to address combinatorial optimization problems. In this study, we leverage the discrete chaotic evolution algorithm to tackle the Traveling Salesman Problem (TSP) for assessment purposes. The chaotic evolution algorithm exploits the ergodicity of chaos to facilitate the search process within the optimization algorithm. It incorporates a mathematical mechanism into the iterative evolution process, simulating ergodic motion within a search space based on a simple principle. To manage the discrete mutation operation within the chaotic evolution algorithm, we introduce a specifically designed chaotic operation. This operation is tailored for its application in solving combinatorial optimization problems. The chaotic sequence plays a crucial role in determining the mutation location. Our evaluation involves the comparison of our proposed discrete chaotic evolution algorithm with the outcomes of the simulated annealing algorithm and the tabu search algorithm. The assessment serves to demonstrate and validate that the discrete chaotic evolution algorithm yields superior optimization performance within the discrete domain.

Breast ultrasound (BUS) image lesion segmentation is an important step in computer-aided diagnosis (CAD) systems for breast cancer screening. Due to acquiring pixel-level labels for fully supervised BUS lesion segmentation being extremely expensive and time-consuming, many studies have adopted weakly supervised methods to mitigate the reliance of models on pixel-level labels. However, in weakly supervised methods, the class activation map (CAM) often excessively focuses on the most distinctive regions of the target while overlooking other lesion parts. This limitation often leads to insufficient CAM activation and compromises the accuracy of lesion segmentation. To address this problem, we propose a weakly supervised framework for lesion segmentation on BUS images using image-level labels. Specifically, the method consists of two stages: classification and segmentation. During the classification stage, the network based on the progressive soft erase (PSE) module reduces the contribution of the most discriminative feature region in CAM, making the network focus on the features of non-dominant regions and generating more high-quality pseudo-masks. In the segmentation stage, we propose a dual-branch feature fusion (DBFF) network, which can not only capture more contextual information by fusing from different scales but also generate a more accurate segmentation mask. Extensive experiments on the publicly available dataset BUS I show the effectiveness of our method. Furthermore, this model significantly supports medical education management by providing a more efficient and cost-effective approach to breast lesion segmentation.

In the social media, users frequently express personal emotions, a subset of which may indicate potential suicidal tendencies. The implicit and varied forms of expression in internet language complicate accurate and rapid identification of suicidal intent on social media, thus creating challenges for timely intervention efforts. The development of deep learning models for suicide risk detection is a promising solution, but there is a notable lack of relevant datasets, especially in the Chinese context. To address this gap, this study presents a Chinese social media dataset designed for fine-grained suicide risk classification, focusing on indicators such as expressions of suicide intent, methods of suicide, and urgency of timing. Seven pre-trained models were evaluated in two tasks: high and low suicide risk, and fine-grained suicide risk classification on a level of 0 to 10. In our experiments, deep learning models show good performance in distinguishing between high and low suicide risk, with the best model achieving an F1 score of 88.39%. However, the results for fine-grained suicide risk classification were still unsatisfactory, with the best weighted F1 score of 50.89%. To address the issues of data imbalance and limited dataset size, we investigated both traditional and advanced, large language model based data augmentation techniques, demonstrating that data augmentation can enhance this model performance by up to 4.65% points in F1-score. Notably, the Chinese MentalBERT model, which was pre-trained on psychological domain data, shows superior performance in both tasks. This study provides valuable insights for automatic identification of suicidal individuals, facilitating timely psychological intervention on social media platforms. The source code and data are publicly available at: https://github.com/HongzhiQ/FineGrainedSuicideDetection.

Acute intracerebral hemorrhage is a life-threatening condition that demands immediate medical intervention. Intraparenchymal hemorrhage (IPH) and intraventricular hemorrhage (IVH) are critical subtypes of this condition. Clinically, when such hemorrhages are suspected, immediate CT scanning is essential to assess the extent of the bleeding and to facilitate the formulation of a targeted treatment plan. While current research in deep learning has largely focused on qualitative analyses, such as identifying subtypes of cerebral hemorrhages, there remains a significant gap in quantitative analysis crucial for enhancing clinical treatments. Addressing this gap, our paper introduces a dataset comprising 222 CT annotations, sourced from the RSNA 2019 Brain CT Hemorrhage Challenge and meticulously annotated at the voxel level for precise IPH and IVH segmentation. This dataset was utilized to train and evaluate seven advanced medical image segmentation algorithms, with the goal of refining the accuracy of segmentation for these hemorrhages. Our findings demonstrate that this dataset not only furthers the development of sophisticated segmentation algorithms but also substantially aids scientific research and clinical practice by improving the diagnosis and management of these severe hemorrhages. Our dataset and codes are available at https://github.com/songchangwei/3DCT-SD-IVH-ICH.

Myasthenia gravis (MG) is a neurological disease that is difficult to diagnose and requires long-term management. The progression of this disease is reflected to some extent in changes in speech, such as hoarseness and articulation disorders. However, it is difficult for general neurologists to grasp the diagnostic patterns of such rare diseases, especially in underdeveloped regions. As an emerging field, speech-based intelligent diagnostic assistance provides a safe, non-invasive, and convenient solution for healthcare education management. To this end, we firstly constructed a novel Chinese speech dataset of myasthenia gravis patients (MGCS). Then we proposed a network named Myasthenia Gravis Speech Net (MGS-Net) for the classification of myasthenia gravis pathological speech, which is mainly composed of two blocks: the Local Feature Enhancement (LFE) block and the Feedforward Dense (FFD) block. The LFE block extracts temporal local features using a sliding window approach, while the FFD block captures the global representation of the data. Compared to existing methods, our pipeline achieves an accuracy of 98.75% and a recall rate of 99.17%. We validated the effectiveness of existing acoustic feature sets in pathological speech classification of MG, which will provide an important tool for health education management of neurological diseases.

The combined models of Graph Neural Network (GNN) and Recurrent Neural Network (RNN) are widely used for patient similarity computation. However, these studies mainly use the medical concepts to organize patient graphs, while a lot of concepts in Electronic Medical Records (EMRs) are paratactic, learning the temporal information based on concept sequences may introduce noise to similarity computation. To address this problem, we propose an Event Graph Learning Network (EGLN) to learn patient similarity. Specially, we firstly leverage the trained Event Extraction (EE) model to obtain the event elements. Then, aggregating the paratactic concepts of each medical event to construct the event graph for the patient to avoid the negative influence of nonexistent temporal information between paratactic concepts. Finally, the spatial and temporal semantic information between event nodes is aggregated for similarity computation. We evaluate EGLN leveraging a real-world dataset, and the experiment results indicate that our proposed EGLN model outperforms all baselines.

Sign language recognition and translation (SLRT) aims to convert sign language into textual representation, which holds significant importance for the deaf community. Sign language possesses complex and diverse grammatical structures, with each sign language having distinct motion trajectories and gesture variations, making SLRT a complex research domain. In recent years, numerous researchers have proposed differ-ent modeling approaches, achieving significant advancements through the utilization of large language models. In this survey, we systematically review the developmental trajectory of SLRT, encompassing an introduction to key technical approaches at each stage and the latest research progress. Through a comprehensive examination of these methods, valuable insights are provided for future research and practical applications. Lastly, we identify the existing limitations of current methods and propose potential avenues for future research.

Integrating lexical knowledge in Chinese named entity recognition (NER) has been proven effective. Among the existing methods, Flat-LAttice Transformer (FLAT) has achieved great success in both performance and efficiency. FLAT performs lexical enhancement for each sentence by constructing a flat lattice (i.e., a sequence of tokens including the characters in a sentence and the matched words in a lexicon) and calculating self-attention with a fully-connected structure. However, the different interactions between tokens, which can bring different aspects of semantic information for Chinese NER, cannot be well captured by self-attention with a fully-connected structure. In this paper, we propose a novel Multi-View Transformer (MVT) to effectively capture the different interactions between tokens. We first define four views to capture four different token interaction structures. We then construct a view-aware visible matrix for each view according to the corresponding structure and introduce a view-aware dot-product attention for each view to limit the attention scope by incorporating the corresponding visible matrix. Finally, we design three different MVT variants to fuse the multi-view features at different levels of the Transformer architecture. Experimental results conducted on four public Chinese NER datasets show the effectiveness of the proposed method. Specifically, on the most challenging dataset Weibo, which is in an informal text style, MVT outperforms FLAT in F1 score by 2.56%, and when combined with BERT, MVT outperforms FLAT in F1 score by 3.03%.

Accurate lesion segmentation on breast ultrasound (BUS) images is a crucial procedure in computer-aided ultrasonic diagnosis. Owing to the privacy of BUS data and the complexity of acquiring pixel-level labels, numerous researches attempt to achieve breast lesion segmentation with predefined feature-based and deep learning-based methods in a unsuper-vised or weakly supervised scenario. Although the former can typically extract more reliable contour information of the lesion, it is severely interfered by irrelevant tissues due to its inability to capture any semantic information. Furthermore, the weakly supervised deep learning segmentation based on class activation map (CAM) can explore the semantic information while failing to provide precise contour information. In view of the above observation, we present a weakly supervised framework merging complementary contour and semantic information for early lesion segmentation in BUS images. Specifically, guided by the prior knowledge of breast anatomy, we first extract and filter the contour information of suspected lesions located in breast parenchyma layer by clustering and morphological characteristic, respectively. Afterward, semantic information extraction is performed by a classification network to automatically explore the category information of the lesion. Finally, we selectively fuse the complementary information to facilitate lesion segmentation performance with more comprehensive features. Extensive experiments are conducted on the public dataset BUSI, and the results confirm the validity of our approach.

We propose a data analysis method that combines the objectives of nonlinear principal component analysis and nonlinear discriminant analysis with the kernel method in a reproducing kernel Hilbert space. This method addresses nonlinear data analysis problems in high-dimensional spaces, specifically the reproducing kernel Hilbert space, through the use of the kernel trick. Our proposed method can be considered as a semi-supervised data analysis approach. We evaluate our proposed method using various kernel functions and datasets, both visually and quantitatively. The evaluation results demonstrate that our proposal outperforms kernel principal component analysis and generalized discriminant analysis in terms of classification performance. This indicates the advantages and originality of our proposed method. Furthermore, we analyze and discuss our findings based on the evaluation results, and highlight potential areas for further research and future work related to our proposal.

Over the past several years, deep learning technologies have made tremendous progress in medical image tasks including classification, segmentation, and object detection. However, there are two main limitations of indocyanine green (ICG) images which are often used in breast cancer related lymphedema (BCRL): insufficient sample numbers and low image quality. Consequently, the conventional deep learning based classification methods such as ResNet have faced challenges in achieving satisfactory results. To tackle the concern, this paper puts forward a deep learning method named Denoising Diffusion Probabilistic Model Self-supervised Knowledge Distillation Net (DDPM-SKDNet) for the ICG images classification task, by incorporating a contrastive learning based approach as the network architecture and using DDPM as the image generator in the contrastive module to expand the dataset size. Furthermore, a knowledge distillation approach is utilized to increase the effectiveness of the network. The proposed method was validated on ICG datasets and achieved a significant improvement in classification accuracy, increasing it from 66.7% in the baseline method to 82.1% in the proposed method.

Biomedical named entity recognition (BNER) is a basic task of the extraction of medical information. The existing deep learning-based approaches usually represent the medical text by using words or characters. However, most of biomedical terms consist of many words (characters). Splitting them into many fragments (words or characters) while leveraging the attention mechanism to assign attention scores for each fragment maybe disperse the importance weight and cause a lower attention score for the biomedical terms. Therefore, this paper presents a dictionary-based adaptive attention network for BNER. Specifically, a biomedical dictionary is firstly constructed by integrating multiple existing medical resources. Secondly, building the guidance vectors by matching the electronic medical record (EMR) text to the constructed dictionary. Then, an adaptive attention strategy is presented to guide the attention mechanism to assign higher attention to the overall medical term by using the guidance vectors. We conduct extensive experiments on a real-world dataset, the results illuminate that our presented method outperforms all baselines.

Understanding gene expression processes necessitates the accurate classification and identification of transcription factors, which is supported by high-throughput sequencing technologies. However, these techniques suffer from inherent limitations such as time consumption and high costs. To address these challenges, the field of bioinformatics has increasingly turned to deep learning technologies for analyzing gene sequences. Nevertheless, the pursuit of improved experimental results has led to the inclusion of numerous complex analysis function modules, resulting in models with a growing number of parameters. To overcome these limitations, it is proposed a novel approach for analyzing DNA transcription factor sequences, which is named as DeepCAC. This method leverages deep convolutional neural networks with a multi-head self-attention mechanism. By employing convolutional neural networks, it can effectively capture local hidden features in the sequences. Simultaneously, the multi-head self-attention mechanism enhances the identification of hidden features with long-distant dependencies. This approach reduces the overall number of parameters in the model while harnessing the computational power of sequence data from multi-head self-attention. Through training with labeled data, experiments demonstrate that this approach significantly improves performance while requiring fewer parameters compared to existing methods. Additionally, the effectiveness of our approach is validated in accurately predicting DNA transcription factor sequences.

Summary Federated edge cloud (FEC) is an edge computing environment where servers in the same edge management domain could collaborate to handle latency‐sensitive services, thus better guaranteeing users' requirements on multiple quality of service (MQoS). Traditional scheduling methods only consider whether the server meets the resource requirements of the service, without paying attention to whether their resource characteristics match. In scenarios where server's resources are dynamically changing, this may reduce the resource utilization and the efficiency of service execution. To address this challenge, a dynamic resource adaptation‐multiple quality of service (DRA‐MQoS) algorithm is proposed for service scheduling in this environment. DRA‐MQoS could dynamically evaluate the resource characteristics of servers and services from the perspectives of “individual” and “overall” by combining the historical scheduling data of services and the utilization of different resources of server clusters. By scheduling the services to servers with the same resource characteristics for execution, the proposed policy fusion algorithm efficiently responds to the dynamically changing quality of service (QoS) demands of users by changing the weight parameters of policies. Simulation results in CloudSimSDN show that the energy consumption and execution time of DRA‐MQoS are reduced by 23% and 12%, respectively, compared with existing methods.

Abstract Triple‐negative breast cancer is the worst prognosis in breast cancer, accounting for 10.0–20.8% of all breast cancers. Considering that triple‐negative breast cancer has great heterogeneity and very poor prognosis, clinical medication guidance is in urgent need of a more detailed classification of breast cancer itself. Although many researchers have been dedicated to the clustering of triple‐negative breast cancer and have found possible targets based on typing, their results are not closely related to the prognosis. This paper utilizes three clustering methods to retype the patient data with triple‐negative breast cancer, and the results show that the triple‐negative breast cancer data could be classified into two categories. Eight important genes and three important clinical factors related to the prognosis of two types of triple‐negative breast cancer have been obtained. These genes have the following three characteristics: co‐expression, differential expression and interaction. In terms of breast cancer control, the prognosis can be controlled as much as possible by regulating gene levels, which provides new directions and ideas for related research on breast cancer prognosis.

Biomedical named entity recognition (BNER) is a critical task for biomedical information extraction. Most popular BNER approaches based on deep learning utilize words and characters as features to represent medical texts. However, many medical terminologies are composed of multiple words and characters, and splitting medical terminology into multiple words (or characters) and assigning weight values for each word (or character) by a standard attention mechanism may disperse the attention score and result in a lower weight value for the medical terminology. This paper proposes a Dictionary-guided Attention Network (DGAN) for BNER in Chinese electronic medical records (EMRs). First, the medical concepts are extracted as large-size words to supplement the comprehensive semantic information of the medical terminology by matching the EMR text to the biomedical dictionary. Then, based on the matched dictionary results, an optimized attention strategy is proposed to focus on the medical concept and adaptively assign higher weights to the characters contained in a concept. Furthermore, semisupervised learning is introduced to reduce the manual labeling of data and to handle the entities not defined in the medical dictionary. To validate our new model in recognizing biomedical named entities, we conduct comprehensive experiments on a real-world Chinese EMR dataset and the CCKS2017 dataset. Our promising results illustrate that our method not only achieves a state-of-the-art performance in BNER but also reduces manual data annotation.

The automatic disease diagnosis utilizing clinical data has been suffering from the issues of feature sparse and high probability of missing values. Since the graph neural network is a effective tool to model the structural information and infer the missing values, it is becoming the dominant method for the predictive model construction from electronic medical records. Existing graph neural network based solutions usually adopt the medical concepts (e.g., symptoms) the feature representation of clinical data without considering their underlying semantic relations. The limited discriminative capability of the medical concept cannot provide sufficient indicative information about the disease. This article proposes a knowledge-guided graph attention network for the disease prediction. Beside extracting the attribute-value structure as a large-size medical concept, the mutual information between multiple medical concepts mentioned in the electronic medical records are taken into account in the graph construction. Meanwhile, the defined diseases and their associations with the medical concepts in the medical knowledge graph are incorporated into the graph, which provides the potentials to enhance the indicative impacts of the medical concepts directly related to a target disease. Then, the spatial and attention based graph encoders are employed to aggregate information from directly neighbor nodes to generate node embeddings as the compact features to be used for disease diagnosis. The approach itself is a general one that can utilized to build the predictive model using Chinese EMRs for different diseases. The empirical experiments for its performance evaluation are conducted on the real-world COPD EMR dataset. The comparison study results show that the proposed model outperforms baseline methods, which illustrates the effectiveness of our proposed model.

Hyperspectral image (HSI) classification attempts to classify each pixel, which is an important means of obtaining land–cover knowledge. Hyperspectral images are cubic data with spectral–spatial knowledge and can generally be considered as sequential data alongside spectral dimension. Unlike convolutional neural networks (CNNs), which mainly focus on local relationship models in images, transformers have been shown to be a powerful structure for qualifying sequence data. However, it lacks the excellent ability of CNNs in establishing local relationships in images and cannot perform good generalization in case of insufficient data. In addition, the gradient disappearance problem hinders the convergence stability of deep learning networks as the layers get deeper. To address these problems, we propose a Cascaded Convolution-based Transformer with Densely Connected Mechanism (CDCformer) for hyperspectral image classification. First, we propose a cascaded convolution feature tokenization to extract spectral–spatial information, which will introduce some inductive bias properties of CNN into the transformer. In addition, we design a simple and effective densely connected transformer to enhance feature propagation and transfer memorable information from shallow to deep layers. It efficiently improves the performance of the transformer and extracts more discriminative spectral–spatial features from the HSI. Extensive experimental evaluation of three public hyperspectral data sets shows that CDCformer achieves competitive classification results.

The effective classification of pollen is a critical method to the prevention of pollen allergy. The conventional pollen classification primarily relies on manual handling under the microscope, which does not only require a lot of manpower but also has low classicization accuracy on results. In this paper, the optical microscope was used to scan the slides and the pollen image dataset was made for classification. We found from the pollen dataset that most of the images have clear contours, but there were also many pollen images with the following problems. One is that the pollen was covered by impurities like dust and pebbles, and the other is that the pollen images stain unevenly and the pollen image was blurred and the number of pollens was unbalanced. This paper proposes a Dual-Channel Attention method, we call it DCANet, which combines the advantages of channel attention and channel self-attention to improve the classification accuracy. The results of classification activation map analysis showed that DCANet paid more attention to pollen images.

Background: Classification of infants has always been considered a crucial task in the literature related to predicting small for gestational age (SGA) infants. Traditional medical guidance becomes increasingly unsatisfactory, as patients' care should be centered not only on clinical symptoms but also on socio-economic and demographic factors. Infants with excessive gestational weight exhibit serious maternal complications that require early intervention to stream-line the progression of the disease. Methods: This research proposes to use the Stacked Generalization Scheme (SGS) to predict Small for Gestational (SGA) Infants on the dataset collected from the National Pre-Pregnancy and Examination Program of China. A Cleaned Feature Vector (CFV) is created that entertains issues related to missing values, discretization of fields, and data imbalance. Later, Knowledge-Driven Data (KDD) Features are extracted from the obtained CFV, and the proposed scheme is applied to predict SGA infants. The proposed scheme superposed the existing baseline approaches by achieving the highest precision, recall, and AUC scores of 0.94, 0.85, and 0.89, respectively. Conclusion: The proposed SGS can predict SGA infants accurately compared to existing baseline schemes using KDD parameters, which can help pediatricians develop an efficient SGA Prognosis process.

To solve the problem that a single Bidirectional Long Short-Term Memory (BILSTM) model can only learn the features of a specific dimension, and the local dependence between characters is easily lost when character-level segmentation is used to extract semantic features, this paper proposes a Chinese medical named entity recognition method based on multi-segmentation and multi-layer BILSTM. The purpose of multi-word segmentation is to enrich the semantic features that can be used in the model learning process, so as to further improve the entity recognition ability of the model. Multi-layer bidirectional long short-term memory can obtain feature information of different dimensions by setting hidden layers of different sizes, and use the attention mechanism to capture global information. Experimental results show that this method can significantly improve the performance of Chinese medical named entity recognition, and obtain 78.69% F value in real electronic medical record dataset.

Microscopic imaging plays an important role in the biomedical field. Existing deep learning based methods rely on high-quality data. However, there is a lot of noise (such as bubbles and impurities) in the microscopic images of biological samples collected outdoors, which may lead to significant interference in the microscopic objects identification task. To solve this problem, this paper proposes a deep learning based method for microscopic object localization and classification. Firstly, the whole slide image is preprocessed to obtain the microscopic images after preliminary filtering bubbles and impurities. Then, the sensitized pollen grains are located based on the deep learning model to remove the interference of remaining impurities, and the microscopic images of sensitized pollen grains are classified. This method can effectively suppress the interference of noise in microscopic images on object classification and improve the accuracy and reliability of model. The proposed method is verified by experiments based on real data and the results show that the proposed method achieves the highest accuracy compared with other deep learning methods.

The medical record system is becoming more and more irreplaceable in the medical industry, and the electronic medical record data continues to grow over time. There is a lot of knowledge and information in these accumulated medical resources that can be used for medical services, but how to obtain this valuable medical information is a difficult problem that needs to be overcome. Event extraction belongs to information extraction technology, which is an effective solution that can automatically mine knowledge and information from text data. Many studies have applied it to the text data of electronic medical records to extract medical events related to medical treatment. They have achieved certain results for medical services. However, these studies usually lack the synergistic consideration of global features and local features of medical text information in terms of Chinese medical record text mining and utilization. To better solve this problem, we try to propose a hybrid neural network model (BCBC) based on CNN-BILSTM-CRF. By integrating CNN and BILSTM, the local and global features of the text are comprehensively extracted, which makes up for the insufficient semantic capture of a single model in the traditional method. Through experimental verification, the hybrid neural network model BCBC proposed in this paper outperforms other previous advanced methods in event extraction and can efficiently complete the event extraction task.

Analyzing character sentiment information through audio is an essential and challenging task. Using convolutional and recurrent neural network approaches for audio sentiment analysis has achieved initial results, proving that deep learning methods can be helpful for sentiment analysis of audio. With the rise of multimodal research, using audio and text feature fusion for sentiment analysis has better results. However, this approach uses the pipeline approach for analysis, which needs to use an additional speech-to-text model to get text information first so that the error of text information conversion will affect the subsequent judgment. In order to solve this problem, we propose a new end-to-end model, which adopts the model architecture of seq2seq, encodes audio information by the encoder, generates text information of audio by the decoder and understands the textual content, and finally obtains the final emotional state by fusing the features extracted by the decoder and the encoder. Our model has experimented on actual psychological assistance hotline data, and the results show that our method is significantly better than the baseline method, which is a meaningful method.

Myopia has become a major public health problem affecting the eye health of our citizens, especially teenagers. Fundus images can be obtained non-invasively and can be used to monitor and follow up on the progress in high myopia. However, with the development of artificial intelligence, it is still difficult to establish a computer-aided diagnosis model for high myopia with young children as research objects, mainly because 1) it is very difficult to collect labeled fundus images, and there is no large amount of such data that can be freely accessed; 2) hard samples which have clinical significance for population screening and diagnosis are rare and indistinguishable. To solve these problems, we propose a high myopia detection model on fundus images based on curriculum learning. We design a dual-curriculum generation module, which aims to use the expert model to endow the curriculum with new training indicators so that the student model can gradually and robustly identify hard samples. Compared with the baseline, our framework significantly improves the convergence speed of the training process and achieves the best performance during testing. Experimental results on a high myopia fundus images dataset show that our framework provides efficient and accurate detection and outperforms other methods.

Pollen classification plays an essential role in many fields such as medicine and palynology. Notably, manual pollen identification via observing key pollen information (e.g., their contours) is time-consuming and laborious. To date, deep learning methods can extract complex features in an end-to-end manner. However, deep learning based automatic classification methods on pollen grains are still rare, and their performances remain unsatisfactory owing to limitation of interference from irrelevant information (such as impurities and bubbles) and the lack of pollen attention. Based on the above considerations, we propose a contour-guided network called CG-Net, which contains three modules. Image pre-processing module first removes impurities and bubbles in pollen images according to the color information. Then, contour awareness module is designed to generate contour features and these features are served as attention maps for next module. Finally, contour guidance module weights the yielded contour attention maps to both original images and feature maps of different convolution layers, making the CNN focus on discriminative features of pollen grains. Extensive experiments are conducted on several real-world pollen datasets, and the results demonstrate the effectiveness of our proposed method with the accuracy and F1-score over 84%.

Face detection tasks under the current epidemic prevention situation often acquire images with partial occlusion. General face detectors ignore the challenge brought by occlusion, making it difficult to meet daily needs. In order to address this problem, this paper proposes a real-time occluded face detection network based on the improved CenterNet with information dropping strategy. First, depth separable convolution and attention mechanism are introduced into the backbone to reduce parameters and extract occlusion-robust features. Second, a feature fusion neck is designed to improve the performance of multi-scale face detection. In addition, the data augmentation method with information removal strategy enriches the diversity of occlusion samples. Experiments indicate that our model improves the fps as well as maintains the accuracy.

Recently, lexicon-based Chinese Named Entity Recognition (NER) models have achieved state-of-the-art performance by benefiting from the rich boundary and semantic information contained in the lexicon. However, in the Chinese medical domain, it’s difficult to obtain the medical lexicon related to the target medical corpus. In this paper, we propose a new paradigm, enhancing Chinese medical NER with Auto-mined Lexicon (ALNER), which alleviates the difficulty of obtaining the medical lexicon by designing a data-driven automatic lexicon construction method. We define medical lexicon construction as a high-quality phrase mining task. We perform secondary annotation on the NER annotated data and use the secondary annotated data to train a deep learning-based phrase tagger. Experimental results show that our method can be combined with different lexicon-based Chinese NER models to improve performance and that the method does not require an external medical lexicon.

Recently, deep neural network (DNN) models work incredibly well, and edge computing has achieved great success in real-world scenarios, such as fault diagnosis for large-scale rotational machinery. However, DNN training takes a long time due to its complex calculation, which makes it difficult to optimize and retrain models. To address such an issue, this work proposes a novel fault diagnosis model by combining binarized DNNs (BDNNs) with improved random forests (RFs). First, a BDNN-based feature extraction method with binary weights and activations in a training process is designed to reduce the model runtime without losing the accuracy of feature extraction. Its generated features are used to train an RF-based fault classifier to relieve the information loss caused by binarization. Second, considering the possible classification accuracy reduction resulting from those very similar binarized features of two instances with different classes, we replace a Gini index with ReliefF as the attribute evaluation measure in training RFs to further enhance the separability of fault features extracted by BDNN and accordingly improve the fault identification accuracy. Third, an edge computing-based fault diagnosis mode is proposed to increase diagnostic efficiency, where our diagnosis model is deployed distributedly on a number of edge nodes close to the end rotational machines in distinct locations. Extensive experiments are conducted to validate the proposed method on the data sets from rolling element bearings, and the results demonstrate that, in almost all cases, its diagnostic accuracy is competitive to the state-of-the-art DNNs and even higher due to a form of regularization in some cases. Benefited from the relatively lower computing and storage requirements of BDNNs, it is easy to be deployed on edge nodes to realize real-time fault diagnosis concurrently.Note to Practitioners—Rotating machines, such as engines and motors, are the cornerstones of the modern industry. Edge computing is an emerging computing paradigm where computation is performed on the edges of networks rather than on the central cloud, thereby reducing system response time, transmission overhead, storage space, and computation resources of the cloud. Motivated by the high demand on computation for deploying DNN models and lower computation complexity for running BDNN models and easiness for large-scale deployment of BDNNs, an edge computing-based method for real-time fault diagnosis of rotating machines is proposed. First, we design a BDNN-based feature extractor to decrease the amount of computation and speed up a diagnosis processes. Then, the resulting binary features are fed to train an RF-based classifier, where we use ReliefF instead of Gini index when training a random forest model to further improve the proposed method’s diagnostic accuracy. Finally, a novel cloud-edge collaborative computing-based fault diagnostic mode is presented, where the model trained from the central cloud is deployed on the edge computing devices distributed in large-scale scenarios to realize real-time fault diagnosis. Experiment results show that the proposed method can maintain the desired accuracy but greatly enhance the diagnosis speed when deployed on the edge nodes near end physical machines. It is easily extended and used for fault detection in many industrial sectors.

Several developments in computational image processing methods assist the radiologist in detecting abnormal breast tissue in recent years. Consequently, deep learning-based models have become crucial for early screening and interpretation of mammographic images for breast masses diagnosis, helping for successful treatment. Breast masses and calcification is an essential parameter for the prognosis of breast cancer. However, the mammographic image’s mass detection needs a deeper investigation due to the breast masses’ heterogeneity and anomalies’ characteristics that are easily confused with other objects present in the image. Hence, this study proposed a deep learning-based convolutional neural network (ConvNet) that will incorporate both mammography and clinical variables to predict and classify breast masses to assist the expert’s decision-making processes. We trained our proposed model with 322 scanned digital mammographic images of the MIAS (Mammogram Image Analysis Society) dataset and 580 images of the private dataset to evaluate the performance, which is highly imbalanced. This study aimed to perform an automatic and comprehensive characterization of breast masses using appropriate layers deep ConvNet model with high accuracy true-positive rate, decreased error rate and applying data-augmentation techniques. We obtained a classification accuracy of 97% applying the filtered deep features, which is the best performance from the existing approaches.

The widely deployed of hospital information systems causes an explosive growth of the electronic medical records (EMRs). It makes the medical structured processing technologies become critical to find researchable data in the large medical dataset. However, the high quality structured processing is a challenging task, in particular due to the inherent complexity and polysemy of medical terminology. In this paper, we propose a novel approach to achieve the joint extraction of events in Chinese electronic medical records, which solves the problem of cascading error transmission in traditional models and the ambiguity of Chinese characters. We first use the Bi-directional Encoder Representation from Transformers(BERT) model to mine features from the preprocessed medical data; then based on the characteristics of Chinese, we use the Bi-directional Long Short-Term Memory(BILSTM) model to capture the semantic information of the context. The experiments were conducted on a real dataset. The F1 score of our model in the identification and classification tasks of event triggers and arguments is the highest, reaching 71.6, 68.1, 55.4 and 46.9, respectively, which proves the effectiveness of the proposed method.

The integration of lexicon information into character-based models is a hot topic in Chinese Named Entity Recognition(NER) research. Most methods only utilize information from a single lexicon which is usually a general lexicon. However, In the Chinese medical text scenario, due to the large amount of medical terminology, a single lexicon, especially a general lexicon, offers little performance improvement to the Chinese NER. In this paper, we propose a Multi-source Lexicon Information Fusion method for Named Entity Recognition in Chinese Medical Text(MLNER) which can utilize information from both general and medical lexicons. Considering the small medical annotated corpus, we combine the model with the pre-trained model to improve the performance of the model on small datasets by exploiting the rich representation capability of the pre-trained model. Experiments show that our method can effectively improve the performance of NER in Chinese medical text. Our model is also applicable to Chinese NER tasks in other domain specific fields, with good scalability and application value.

Cataract is a chronic eye disease that causes irreversible vision loss. Automatic cataract detection can help people prevent visual impairment and decrease the possibility of blindness. To date, many studies utilize deep learning methods to grade cataract severity on fundus images. However, they mainly focus on the classification performance and ignore the model interpretability, which may lead to a semantic gap between networks and users. In this paper, we present a deep learning network to improve the model interpretability, which consists three main modules: deep feature extraction, visual saliency module and semantic description module. Visual and semantic interpretation jointly employed to provide cataract-grade oriented interpretation for the overall model. Experimental results on real clinical data set show that our method improves the interpretability for cataract grading while ensuring the high classification performance.

The wide adoption of electronic medical record (EMR) systems causes rapid growth of medical and clinical data. It makes the medical named entity recognition (NER) technologies become critical to find useful patient information in the medical dataset. However, the medical terminologies usually have the characteristics of inherent complexity and ambiguity, it is difficult to capture context-dependency representations by supervision signal from a simple single layer structure model. In order to address this problem, this paper proposes a hybrid model based on stacked Bidirectional Long Short-Term Memory (BILSTM) for medical named entity recognition, which we call BSBC (BERT combined with stacked BILSTM and CRF). First, we use Bidirectional Encoder Representation from Transformers (BERT) to perform unsupervised learning on an unlabeled dataset to obtain character-level embeddings. Then, stacked BILSTM is utilized to obtain context-dependency representations through the multi hidden layers structure. Finally, Conditional Random Field (CRF) is used to predict sequence tags. The experiment results show that our method significantly outperforms the baseline methods, it serves as a strong alternative approach compared with traditional methods.

Hydronephrosis may lead to many potential diseases, and the diagnosis of hydronephrosis is time-consuming and laborious. To assist physicians in hydronephrosis diagnosis and treatment planning, an accurate and automatic kidney segmentation method is highly required in clinical practice. In recent years, deep convolutional neural networks such as Unet plays a key role in the field of image segmentation, but Unet itself cannot adjust the receptive field actively, which may result in poor attention to the characteristics of the segmented target. We propose an encoder-decoder network with weighted skip connections and the idea of hierarchical equal resolution that can manually control the receptive field. We evaluated our method by comparing it with various classical networks using a dataset of 1850 annotated images. The MPA of the model is 94.12 and the MIoU is 89.49, which outperformed other classical networks we compared to.

Hydronephrosis is a common renal disease in children which can lead to a series of complications, and ultrasonography is a basic examination usually performed on suspected hydronephrosis patients. If we can use deep learning approaches to judge and grade the disease in the ultrasonic examination stage, we can save a lot of manpower, medical resources, money, and help the suffered patients. For the semantic segmentation of kidney ultrasound image, we designed an Attention-based Pyramid Scene Parsing Network (A-PSPNet), the core of which is the basic feature extraction network combining Convolutional Block Attention Module (CBAM) and pyramid analysis module. Experiments were carried out on a hydronephrosis dataset containing 1850 annotated ultrasound images, including the arrangement of attention units, statistical computing power, and comparison of the effectiveness between the benchmark and our proposed method. Our constructed model achieved better segmentation performance than benchmarks with only little extra overhead, which validated the lightweight and effectiveness of the model.

With the rapid increase of text data generated by the Internet, the problem of text information overload is becoming more and more serious. Automatic text summarization provides a good method to simplify text information. Traditional methods are mainly divided into extractive and abstractive methods. However most extractive methods do not have too much context connection, which leads to unsmooth abstracts. Abstractive method is the mainstream method, but it also deviates from the text content and has the problem of poor readability. In this paper, a hybrid automatic text summarization method is proposed based on deep learning and a rapid self-attention mechanism. This mechanism is used to obtain a hybrid model DNM (Dilated Neural Random Attention with Minimal Risk Loss) by optimizing the network structure and combining it with a specific loss function. The ROUGE score of our model is compared with the extractive Neural Document Summarization (NEUSUM), the abstractive Graph-Based Attentional (GBA) and the hybrid model CopyNet on the LCSTS dataset so as to achieve more accurate and reasonable automatic text summarization.

Network forensics can be an expansion associated with network security design which typically emphasizes avoidance and detection of community assaults. It covers the necessity for dedicated investigative abilities. When you look at the design, this indeed currently allows investigating harmful behavior in communities. It will help organizations to examine external and community this is undoubtedly around. It is also important for police force investigations. Network forensic techniques can be used to identify the source of the intrusion and the intruder’s location. Forensics can resolve many cybercrime cases using the methods of network forensics. These methods can extract intruder’s information, the nature of the intrusion, and how it can be prevented in the future. These techniques can also be used to avoid attacks in near future. Modern network forensic techniques face several challenges that must be resolved to improve the forensic methods. Some of the key challenges include high storage speed, the requirement of ample storage space, data integrity, data privacy, access to IP address, and location of data extraction. The details concerning these challenges are provided with potential solutions to these challenges. In general, the network forensic tools and techniques cannot be improved without addressing these challenges of the forensic network. This paper proposed a thematic taxonomy of classifications of network forensic techniques based on extensive. The classification has been carried out based on the target datasets and implementation techniques while performing forensic investigations. For this purpose, qualitative methods have been used to develop thematic taxonomy. The distinct objectives of this study include accessibility to the network infrastructure and artifacts and collection of evidence against the intruder using network forensic techniques to communicate the information related to network attacks with minimum false-negative results. It will help organizations to investigate external and internal causes of network security attacks.

With rapid industrial development, air pollution problems, especially in urban and metropolitan centers, have become a serious societal problem and require our immediate attention and comprehensive solutions to protect human and animal health and the environment. Because bad air quality brings prominent effects on our daily life, how to forecast future air quality accurately and tenuously has emerged as a priority for guaranteeing the quality of human life in many urban areas worldwide. Existing models usually neglect the influence of wind and do not consider both distance and similarity to select the most related stations, which can provide significant information in prediction. Therefore, we propose a Geographic Self-Organizing Map (GeoSOM) spatiotemporal gated recurrent unit (GRU) model, which clusters all the monitor stations into several clusters by geographical coordinates and time-series features. For each cluster, we build a GRU model and weighted different models with the Gaussian vector weights to predict the target sequence. The experimental results on real air quality data in Beijing validate the superiority of the proposed method over a number of state-of-the-art ones in metrics, such as${R} ^{2}$, mean relative error (MRE), and mean absolute error (MAE). The MAE, MRE, and${R} ^{2}$are 16.1, 0.79, and 0.35 at the Gucheng station and 19.53, 0.82, and 0.36 at the Dongsi station.

Atmospheric visibility is an indicator of atmospheric transparency and its range directly reflects the quality of the atmospheric environment. With the acceleration of industrialization and urbanization, the natural environment has suffered some damages. In recent decades, the level of atmospheric visibility shows an overall downward trend. A decrease in atmospheric visibility will lead to a higher frequency of haze, which will seriously affect people's normal life, and also have a significant negative economic impact. The causal relationship mining of atmospheric visibility can reveal the potential relation between visibility and other influencing factors, which is very important in environmental management, air pollution control and haze control. However, causality mining based on statistical methods and traditional machine learning techniques usually achieve qualitative results that are hard to measure the degree of causality accurately. This article proposed the seq2seq-LSTM Granger causality analysis method for mining the causality relationship between atmospheric visibility and its influencing factors. In the experimental part, by comparing with methods such as linear regression, random forest, gradient boosting decision tree, light gradient boosting machine, and extreme gradient boosting, it turns out that the visibility prediction accuracy based on the seq2seq-LSTM model is about 10% higher than traditional machine learning methods. Therefore, the causal relationship mining based on this method can deeply reveal the implicit relationship between them and provide theoretical support for air pollution control.

Large for gestational (LGA) means the fetus having an abnormal birth weight. It adheres severe complications during and after the maternal period. Therefore, this research presents an ensemble classification scheme using Chinese National Pre-Pregnancy Examination Program dataset to classify a fetus as an LGA or non-LGA based on provided Chinese LGA classification guidelines. Moreover, the proposed scheme is comprised of data cleansing and ensemble classification schemes that have drastically improved the LGA classification process with improved performance results compared to present published studies. Therefore, the recommended scheme can be utilized by healthcare professionals to build an enhanced and reliable LGA classification system.

Recently, the automatic diagnosis of Turner syndrome (TS) has been paid more attention. However, existing methods relied on handcrafted image features. Therefore, we propose a TS classification method using unsupervised feature learning. Specifically, first, the TS facial images are preprocessed including aligning faces, facial area recognition and processing of image intensities. Second, pre-trained convolution filters are obtained by K-means based on image patches from TS facial images, which are used in a convolutional neural network (CNN); then, multiple recursive neural networks are applied to process the feature maps from the CNN to generate image features. Finally, with the extracted features, support vector machine is trained to classify TS facial images. The results demonstrate the proposed method is more effective for the classification of TS facial images, which achieves the highest accuracy of 84.95%.

BACKGROUND: Screening of the brain computerised tomography (CT) images is a primary method currently used for initial detection of patients with brain trauma or other conditions. In recent years, deep learning technique has shown remarkable advantages in the clinical practice. Researchers have attempted to use deep learning methods to detect brain diseases from CT images. Methods often used to detect diseases choose images with visible lesions from full-slice brain CT scans, which need to be labelled by doctors. This is an inaccurate method because doctors detect brain disease from a full sequence scan of CT images and one patient may have multiple concurrent conditions in practice. The method cannot take into account the dependencies between the slices and the causal relationships among various brain diseases. Moreover, labelling images slice by slice spends much time and expense. Detecting multiple diseases from full slice brain CT images is, therefore, an important research subject with practical implications. RESULTS: In this paper, we propose a model called the slice dependencies learning model (SDLM). It learns image features from a series of variable length brain CT images and slice dependencies between different slices in a set of images to predict abnormalities. The model is necessary to only label the disease reflected in the full-slice brain scan. We use the CQ500 dataset to evaluate our proposed model, which contains 1194 full sets of CT scans from a total of 491 subjects. Each set of data from one subject contains scans with one to eight different slice thicknesses and various diseases that are captured in a range of 30 to 396 slices in a set. The evaluation results present that the precision is 67.57%, the recall is 61.04%, the F1 score is 0.6412, and the areas under the receiver operating characteristic curves (AUCs) is 0.8934. CONCLUSION: The proposed model is a new architecture that uses a full-slice brain CT scan for multi-label classification, unlike the traditional methods which only classify the brain images at the slice level. It has great potential for application to multi-label detection problems, especially with regard to the brain CT images.

Summary The motif discovery of multi‐dimensional time series datasets can reveal the underlying behavior of the data‐generating mechanism and reflect the relationship between time series in different dimensions. The study of motif discovery is of important significance in environmental management, financial analysis, healthcare, and other fields. With the growth of various information acquisition devices, the number of multi‐dimensional time series datasets is rapidly increasing. However, it is difficult to apply traditional multi‐dimensional motif discovery methods to large‐scale datasets. This paper proposes a novel method for motif discovery and analysis in large‐scale multi‐dimensional time series. It can effectively find multi‐dimensional motifs and the correlation among the motifs. The experimental results show that the proposed method achieves better performance than the related arts on synthetic and real datasets. It is further validated on practical air quality data and provides theoretical support for real air pollution control in places such as Beijing.

Diagnosing infants who are small for gestational age (SGA) at early stages could help physicians to introduce interventions for SGA infants earlier. Machine learning (ML) is envisioned as a tool to identify SGA infants. However, ML has not been widely studied in this field. To develop effective SGA prediction models, we conducted four groups of experiments that considered basic ML methods, imbalanced data, feature selection and the time characteristics of variables, respectively. Infants with SGA data collected from 2010 to 2013 with gestational weeks between 24 and 42 were detected. Support vector machine (SVM), random forest (RF), logistic regression (LR) and Sparse LR models were trained on 10-fold cross validation. Precision and the area under the curve (AUC) of the receiver operator characteristic curve were evaluated. For each group, the performance of SVM and Sparse LR was similarly well. LR without any sparsity penalties performed worst, possibly caused by the overfitting problem. With the combination of handling imbalanced data and feature selection, the RF ensemble classifier performed best, which even obtained the highest AUC value (0.8547) with the help of expert knowledge. In other cases, RF performed worse than Sparse LR and SVM, possibly because of fully grown trees.

Cataract is one of the most serious eye diseases leading to blindness. Early detection and treatment can reduce the rate of blindness in cataract patients. However, the professional knowledge of ophthalmologists is necessary for the clinical cataract detection. Therefore, the potential costs may make it difficult for the widespread use of cataract detection to prevent blindness. Artificial intelligence assisted diagnosis based on medical images has attracted more and more attention of researchers. Many studies have focused on the use of pre-defined feature sets for cataract classification, but the predefined feature sets may be incomplete or redundant. On account of the aforementioned issues, some studies have proposed deep learning methods to automatically extract image features, but all based on global features and none has analyzed the layer-by-layer transformation process of the middle-tier features. This paper uses convolutional neural networks (CNN) to learn useful features directly from input data, and deconvolution network method is employed to investigate how CNN characterizes cataract layer-by-layer. We found that compared to the global feature set, the detail vascular information, which is lost after multi-layer convolution calculation also plays an important role in cataract grading task. And this finding fits with the morphological definition of fundus image. Through the finding, we gained insights into the design of hybrid global-local feature representation model to improve the recognition performance of automatic cataract grading.

Identifying if a subsurface target is salt or not automatically and accurately is of vital importance to oil drilling. But unfortunately, obtaining the precise position of large salt deposits is very difficult. Professional seismic imaging still requires the interpretation of salt bodies by experts. This leads to very subjective, highly variable renderings. More alarmingly, it leads to potentially dangerous situations for drillers in oil and gas companies. In this paper, a Squeeze-Extraction Feature Pyramid Networks (referred to as Se-FPN) was proposed to tackle the task of image segmentation of salt deposits. Specifically, we utilized SeNet as backbone so as to implicitly learn to suppress irrelevant regions in an input image while highlighting salient features useful for the task. Considering the importance of multi-scales information, we proposed an improved FPN to integrate information of different scales. In order to further fuse the information from multiple scales, the Hypercolumns module was inserted at the end of the network. The proposed Se-FPN has been applied to the TGS Salt Identification Challenge and achieved high quality segmentation effect. The Mean Intersection over Union value can reach 0.86.

Exploring the temporal relationship among events in patient electronic medical records (EMR) is an important problem in biomedical informatics and the results can reveal patients' impending disease conditions. In this paper, we investigate the problem of mining patterns from a sequence of point events, i.e., we only have the information on when the event happens but no duration or numerical value available. We propose a whole pipeline, including event preprocessing, pattern mining, and outcome analysis to mine the patterns and evaluate their effectiveness and discriminative power. Finally, we treat those mined patterns as additional features and evaluate them in a predictive modeling task for the early detection of congestive heart failure. On a real-world EMR data warehouse, we found that by adding those sequential pattern features, the prediction performance could be significantly improved approximately 0.1.

Graph clustering has attracted much attention in recent years, which has wide applications in social and biological networks. Recent approaches on graph clustering mainly focus on either certain graphs with node attributes or uncertain graphs without node attributes. However, many real-world graphs have both uncertainty on the edges and attributes on the nodes. We refer to such networks as \emph{attributed uncertain graphs}. Different from conventional graphs, attributed uncertain graphs post two major challenges for graph clustering: 1) uncertainty on the edges, which makes it difficult to extract reliable clusters; 2) high dimensional attributes on the nodes, which contain irrelevant and noisy information. In this paper, we study the problem of node clustering on attributed uncertain graphs, where we exploit both the uncertain edges and a set of important attributes for graph clustering. The uncertain edges can help identify the set of relevant attributes in the nodes, which are called focus attributes. While the focus attributes can help reduce the uncertainty in edges for graph clustering. We propose two novel approaches: AUG-I based upon integrated attribute induced graphs and AUG-U based upon the unified partition over possible worlds of a uncertain graph. Extensive empirical studies on real-world datasets demonstrate the effectiveness of our approaches for clustering tasks on attributed uncertain graphs.

Retinal fundus image can perceive deep-seated blood vessels in the human body in a non-invasive manner. Retinal blood vessels are the primary anatomical structure that can be visible in the fundus image, while changes in the structural feature of retinal blood vessels cannot only reflect all sort of pathological changes but also serve as an important evidence for diagnosing cataract and other diseases. Automatic fundus image processing and analyzing in the computer has a significant effect on the auxiliary medical diagnosis. Moreover, the blood vessels extracted can be used as a feature for the classification of cataract fundus images. Most of the blood vessel extraction methods often used a heuristic feature set that are usually be extracted manually. For the limitations of current methods, we propose to use deep learning to identify blood vessels, which can perform automatic feature learning. We collected the dataset containing fundus images of 5620 patients for the extraction of blood vessels. We then performed Preprocessing by extracting green channel components and histogram equalization. We also present FCN structure in the fusion of dual sources in which preprocessed grayscale image and the edge information processed by the Sobel operators are used as an input. We also document that FCN enhance the richness of the input features and improve the accuracy. It can be concluded that the proposed method achieves the optimal accuracy for recognizing blood vessels of patients with cataract. Moreover, the accuracy of extracting normal fundus vessels reaches 94.91%. Furthermore, we are intended to use this proposed method for the vascular identification of other medical images.

Presents the introductory welcome message from the conference proceedings. May include the conference officers' congratulations to all involved with the conference event and publication of the proceedings record.

Air pollution is threatening human's health since the industrial revolution, but there are not efficient ways to solve air pollution, so forecasting air quality has become an efficient measure to prevent citizens from hurting of heavy air pollution. In this paper, we proposed an advanced Seq2Seq (Sequence to Sequence) model called attention-based air quality forecasting model (ABAFM) whose RNN encoder is replaced by pure attention mechanism with position embedding. This improvement not only reduces the training time of Seq2Seq model with attention but also enhances the robustness of Seq2Seq models. We implemented ABAFM in Olympic center and Dongsi monitoring stations in Beijing to forecast PM2.5 in future 24 hours. The experimental results showed that the proposed model outperformed the related arts, especially in sudden changes.

Cataract is defined as a lenticular opacity presenting usually with poor visual acuity. It is considered the most common cause of blindness. Early diagnosis and treatment can reduce the suffering of patients and prevent visual impairment from turning into blindness. Recently, cataract diagnosis applying pattern recognition is in a rising period. For retinal fundus images, the task is usually cataract classification. However, it needs complex manual processing, which demands dexterous people and time taking exertion. Besides, it faces the challenge of effective interpretability and dependability. In this paper, we develop a deep-learning algorithm to intuitively identify cataract attributes to solve these limitations. Our model, is a 18(50)-layer convolutional neural network that inputs retinal image in G channel and outputs the prediction with heatmap. The heatmap localizes the areas where most indicative of different levels of cataract. Furthermore, we extend the training strategy for the corresponding task, which aims at improving the performance of the network. Comparing with other methods in cataract classification, we succeeded to achieve state of the art accuracy of proposed method on detection and grading task. Most importantly, our model provides a compelling reason via localizing the areas revealing cataract in the image.

The diagnosis of breast cancer in the middle and early period is conducive to later treatment, but the current diagnosis rate is not very desirable. Using machine learning to predict the benign and malignant of breast cancer can provide some assist to doctors' treatment in clinical practice. In this paper, we have collected data from digitized images of a fine needle aspirate (FNA) of a breast mass. They describe characteristics of the cell nuclei presented in the image. This work adopts several feature selection methods to select the most related features for breast cancer diagnosis. Based on the selected features, four machine learning models, Support Vector Machine (SVM), Decision Tree (DT), AdaBoost and Random Forest (RF) are built and their performance are evaluated. The experimental results show that the accuracy of RF is higher than the other three methods.

The concept of Social Internet of Things (SIoT) has emerged from the integration of social networking into the core of the Internet of Things (IoT). It envisions IoT objects and devices to have social interactions with each other autonomously, cooperate with other agents, and exchange information with human users and surrounding computing devices. These objects are able to sense/actuate, store, and interpret information in an opportunistic and loosely coupled fashion. The objects in the SIoT paradigm can exhibit multiple forms of social relationships derived from their collaborative activities or functional, temporal and spatial dependencies to meet a particular need of human users, which signify the difference between the SIoT domain to that of social-based mobile networks or sensor networks. The social interaction among the SIoT objects contribute a huge volume of data to be processed and used by various applications such as social VANET, social connected health, SIoT-based recommendation service, traffic service, policing, energy management etc, in the area of Smart Cities, Smart Homes, Smart Grid, and Smart Factories to satisfy human needs, interests, and objectives. Such a dynamic landscape with billions of social communities of objects and devices requires new models, theories, and approaches of interaction and collaboration, which could be established by referring to the experience that people have already gained in social networking domain over the past few years.

It is well known that excessive drinking could result in human diseases (e.g., liver disease and cardiovascular disease). To investigate the underlying mechanisms of the effects of alcohol consumption, dental pulp stem cells have been used due to their self-renewal capability, multi-lineage differentiation, and clonogenic efficiency. Researchers usually focus on detecting genes having different mean expression levels between different conditions. However, the variance of gene expression might also provide useful information. In this study, we applied the Brown. Forsythe (BF) test for equal variance based on a human dental pulp stem cells dataset to explore the role of alcohol on the variation of gene expression. We identified two differential variable (DV) gene probes, 213597_s_at near gene CTDSPL (FDR-adjusted p-value = 1.50x10-5) and 213993_at near gene SPON1 (FDR-adjusted p-value=0.042). To best of our knowledge, no studies have found yet the associations of CTDSPL and/or SPON1 to alcohol consumption. GeneMANIA showed that these two genes are related to 20 genes. Web based gene set analysis toolkit (WebGestalt) showed that these 22 genes are enriched in 4 alcohol-related KEGG pathways: adherens junction, TGF-beta signaling pathway, signaling pathways regulating pluripotency of stem cells, and Hippo signaling pathway.

Moderate alcohol consumption has health benefits of adults, while excessive alcohol has poisonous effects on numerous different organs. However, the underlying molecular mechanisms of this phenomenon are still largely unclear yet. Human dental pulp stem cells (hDPSCs) have been used as models to investigate the effect of alcohol (EtOH) on gene expression. It has been demonstrated that both EtOH dosage and exposure time affects gene expression. However, the investigation into whether there exists synergistic effect of EtOH dosage and exposure time on gene expression is rare. In the present study, we reported that the effect of interaction between EtOH dosage and exposure time is significantly associated with the expression of genes SLC25A36 and ZFAND5 based on an hDPSC experiment. This result is novel in that genes SLC25A36 and ZFAND5 have not been reported yet to be associated with alcohol consumption in the existing literature.

Dental pulp stem cells (DPSCs) are stem cellspresent in the dental pulp and have same renewal anddifferentiating properties as bone marrow cells. Hence, DPSCs canbe used for future regenerative therapies of various diseases.Previous studies show that DPSCs would lose their differentiationcapability during long-term passage when they were cultured inmedium promoting differentiation. Researchers have beenworking to explore the key genes that influence their differentiation capability. However, the differentiation of dental pulp stem cells is accompanied by a complex biological process involving multiple genes and multi-element interactions. In this study, we used the weighted gene co-expression network (WGCNA) to obtain gene network modules based on a gene microarray data set from a DPSCs experiment, which is available at the Gene Expression Omnibus (GEO) public database. We identified two modules (yellow module and salmon module) significantly associated with DPSC passage.

Electronic medical records (EMRs) have high value for research, as they contain the patient's personal information, medical history, clinical examination, treatment process, and other information. Analysis based on EMRs can effectively assist doctors in clinical decision-making, provide data support for clinical research as well as personalized healthcare service for patients. We introduce a novel approach for EMR similarity computation by re-structuring and filtering some parts of physical examination result. Our approach is motivated by observations that it is easier to distinguish disease bias special part than bias the whole EMR which maybe contain some ineffective information. Assuming the check parts are independent, we split them and select effective parts. Then, we apply Deep NLP, converting the word to vectors which can be used to measure syntactic and semantic word similarities better. In addition, We replace traditional Euclidean distance with Word Mover's Distance(WMD), a novel distance function between text documents. Finally, KNN cluster is been used to evaluate the similarity between EMRs. Compared with traditional method such as LDA and LSI, our proposed method achieved higher recall value of disease classification problem.

Oral squamous cell carcinomas (OSCC) is the most common head and neck cancer worldwide, with more than 300,000 new cases being diagnosed annually. Studies have shown that miRNAs are involved in the process of growth, differentiation, apoptosis, invasion and metastasis of OSCC tumor cells. How miRNAs work together to contribute to this process is still largely unknown. The goal of our study was to characterize the coexpression network of miRNAs and to identify the miRNA subnetworks (modules) that were significantly associated with the OSCC cancer status. We also searched hub miRNAs that might play a vital role in the development of OSCC. We applied the weighted gene co-expression network analysis (WGCNA) to the miRNA expression profile data from a paired design study contributed by Shiah et al. To account for the within-pair correlation, a linear mixed model (LMM) was constructed to test the associations of miRNA modules to cancer status. Two significant modules (turquoise module with 254 miRNAs and grey module with 309 miRNAs) were identified. The miRNA miR-let-7c was the hub miRNA in the turquoise module in terms of node degree. Finally, we used miRsystem to perform the target gene prediction and KEGG pathway enrichment analysis of miRNAs within the two modules. Interestingly, the two modules have similar sets of target genes so that the top 6 enriched KEGG pathways for the 2 modules were the same. Compared with the probe-wise test used by Shiah et al., we took the network approach and identified significant OSCC-associated miRNA modules, which could help uncover the mechanism that miRNAs interplay each other to contribute to OSCC.

Detection and recognition of textual information in scene images is useful but challenging tasks. Numerous methods have been proposed to solve the problem. Recently the best results are attained by deep neural network based methods. Training such networks needs large amounts of bounding box-level or pixel-level annotated data. Generating large amounts of such data always requires huge amounts of labor which can be expensive and time consuming. In this paper we explore the utilization of weakly supervised deep neural network for generating text proposals in natural scene images. The network allows multi-scale inputs and is trained to perform whole image binary classification to tell whether an image contains text or not. After training the network acquired learning of powerful discriminated features that are capable of distinguishing text from other objects. To get the text location, text confidence score map is generated based on feature maps from the top two convolutional layers by extracting class activation map. Value of each pixel in the score map denotes the confidence score of whether the pixel belongs to text or not. By setting a threshold the score map is converted to a binary mask map. Foregrounds of the mask map are probable text areas. Then Maximally Stable Extremal Regions (MSERs) are extracted from these probable text areas and are aggregated as groups. By processing these groups, text proposals are obtained. Experimental results show that without using any bounding boxes or pixel-level annotation, the algorithm achieves recall rate comparable to some fully supervised methods in ICDAR 2013 focused text dataset and In ICDAR 2015 incidental text dataset.

At present, high road surface temperature (RST) is threatening the safety of expressway transportation. It can lead to accidents and damages to road, accordingly, people have paid more attention to RST forecasting. Numerical methods on RST prediction are often hard to obtain precise parameters, whereas statistical methods cannot achieve desired accuracy. To address these problems, this paper proposes GBELM-RBF method that utilizes gradient boosting to ensemble Radial Basis Function Extreme Learning Machine. To evaluate the performance of the proposed method, GBELM-RBF is compared with other ELM algorithms on the datasets of airport expressway and Badaling expressway during November 2012 and September 2014. The root mean squared error (RMSE), accuracy and Pearson Correlation Coefficient (PCC) of these methods are analyzed. The experimental results show that GBELM-RBF has the best performance. For airport expressway dataset, the RMSE is less than 3, the accuracy is 78.8% and PCC is 0.94. For Badaling expressway dataset, the RMSE is less than 3, the accuracy is 81.2% and PCC is 0.921.

For flexible manufacturing systems, this paper presents a Petri net based deadlock prevention method to obtain highly permissive and structurally minimal supervisors in the presence of uncontrollable and unobservable transitions. First, we define admissible markings which form a maximal strongly connected component (SCC) containing the initial marking and not leading outside the SCC uncontrollably. Then, an integer linear programming problem is formulated to construct a supervisor which permits all admissible markings and forbids all the very first inadmissible states from the admissible zone. It also ensures that no uncontrollable transitions are controlled and no unobservable ones are observed by the supervisor. The method can be applied to the plant nets whose crucial transitions are uncontrollable and/or unobservable. In addition, the obtained supervisors are deadlock-free, highly permissive, and structurally minimal in the presence of controllability and observability.

Medical search technologies are crucial to enable the user to rapidly and effectively discover useful information from massive medical and clinical data. Because of the complexity of medical terminology, traditional information search methods have not fully expressed the intention of the query request and explored the potential semantic knowledge in the document. In this paper, we propose a multi-analysis approach by considering the medical ontology as a semantic resource, which can excavate latent semantic information of a user's query request. In addition, we also recognize topics of medical documents to express text contents for providing support for calculating the similarity between query keywords and documents. Our experiments on PubMed medical article collections show that the semantic-based multi-analysis approach is feasible and efficient compared with other traditional approaches in medical retrieval.

The scale of the modern city has been expanding, which leads to a lot of serious environmental pollution problems. Among them, the air pollution problem is the most prominent. In order to control the air pollution in urban cities, the government has deployed a lot of air pollutant monitoring equipment which produce massive multi-dimensional time series data. Through the motif discovery and analysis of these multi-dimensional time series, we can find the relationships and the rules between air pollutants to provide support and suggestions for controlling the air pollution. In this paper, a novel method on motif discovery and analysis for large-scale multi-dimensional time series data is proposed. The new method can effectively find multi-dimensional motifs and the correlation between them as much as possible, which reveals the underlying rule of different air pollutants. It is validated on practical historical data of air pollutants in Beijing. The experimental results show that the proposed method could obtain better performance than the related work.

A key factor of win–win cloud economy is how to trade off between the application performance from customers and the profit of cloud providers. Current researches on cloud resource allocation do not sufficiently address the issues of minimizing energy cost and maximizing revenue for various applications running in virtualized cloud data centers (VCDCs). This paper presents a new approach to optimize the profit of VCDC based on the service-level agreements (SLAs) between service providers and customers. A precise model of the external and internal request arrival rates is proposed for virtual machines at different service classes. An analytic probabilistic model is then developed for non-steady VCDC states. In addition, a smart controller is developed for fine-grained resource provisioning and sharing among multiple applications. Furthermore, a novel dynamic hybrid metaheuristic algorithm is developed for the formulated profit maximization problem, based on simulated annealing and particle swarm optimization. The proposed algorithm can guarantee that differentiated service qualities can be provided with higher overall performance and lower energy cost. The advantage of the proposed approach is validated with trace-driven simulations.

The recent development of microarray gene expression techniques have made it possible to offer phenotype classification of many diseases. However, in gene expression data analysis, each sample is represented by quite a large number of genes, and many of them are redundant or insignificant to clarify the disease problem. Therefore, how to efficiently select the most useful genes has been becoming one of the most hot research topics in the gene expression data analysis. In this paper, a novel unsupervised two-stage coarse-fine gene selection method is proposed. In the first stage, we apply the kmeans algorithm to over-cluster the genes and discard some redundant genes. In the second stage, we select the most representative genes from the remaining ones based on matrix factorization. Finally the experimental results on several data sets are presented to show the effectiveness of our method.

The economy of scale provided by cloud attracts a growing number of organizations and industrial companies to deploy their applications in cloud data centers (CDCs) and to provide services to users around the world. The uncertainty of arriving tasks makes it a big challenge for private CDC to cost-effectively schedule delay bounded tasks without exceeding their delay bounds. Unlike previous studies, this paper takes into account the cost minimization problem for private CDC in hybrid clouds, where the energy price of private CDC and execution price of public clouds both show the temporal diversity. Then, this paper proposes a temporal task scheduling algorithm (TTSA) to effectively dispatch all arriving tasks to private CDC and public clouds. In each iteration of TTSA, the cost minimization problem is modeled as a mixed integer linear program and solved by a hybrid simulated-annealing particle-swarm-optimization. The experimental results demonstrate that compared with the existing methods, the optimal or suboptimal scheduling strategy produced by TTSA can efficiently increase the throughput and reduce the cost of private CDC while meeting the delay bounds of all the tasks.

In recent years, air quality has become a severe environmental problem in China. Since bad air quality brought significant influences on traffic and people's daily life, how to predict the future air quality precisely and subtly, has been an urgent and important problem. In this paper, a Spatio-Temporal Extreme Learning Machine (STELM) method is proposed for air quality prediction. STELM considers temporal and spatial characteristics of air quality data and related meteorological data, constructs a prediction model based on ELM, and realizes air quality prediction with more than 80% precision. A prototype system is implemented and the experiments on practical air quality data in Beijing validate the effectiveness of our method and system.

In multi-view 3D object retrieval, each object is characterized by a group of 2D images captured from different views. Rather than using hand-crafted features, in this paper, we take advantage of the strong discriminative power of convolutional neural network to learn an effective 3D object representation tailored for this retrieval task. Specifically, we propose a deep embedding network jointly supervised by classification loss and triplet loss to map the high-dimensional image space into a low-dimensional feature space, where the Euclidean distance of features directly corresponds to the semantic similarity of images. By effectively reducing the intra-class variations while increasing the inter-class ones of the input images, the network guarantees that similar images are closer than dissimilar ones in the learned feature space. Besides, we investigate the effectiveness of deep features extracted from different layers of the embedding network extensively and find that an efficient 3D object representation should be a tradeoff between global semantic information and discriminative local characteristics. Then, with the set of deep features extracted from different views, we can generate a comprehensive description for each 3D object and formulate the multi-view 3D object retrieval as a set-to-set matching problem. Extensive experiments on SHREC'15 data set demonstrate the superiority of our proposed method over the previous state-of-the-art approaches with over 12% performance improvement.

Cataract is a lens opacification caused by protein denaturation which leads to a decrease in vision and even results in complete blindness at later stages. The concept of a classification system of automatic cataract detection based on retinal fundus images has been proposed in previous research work which consists of fundus image preprocessing, feature extraction and the building of classifier. This paper proposes to make use of the method of PCA (principal component analysis) to reduce the dimensionality of two sets of features extracted from fundus images which are wavelet features and sketch features, respectively. We find the classification accuracy rate based on new features after PCA transformation is nearly the same with the original ones, but the computation cost can really be decreased a lot. Experiment results provide a bright foresight in later practical application of classification system of automatic cataract detection.

Background modeling is a fundamental problem in computer vision and usually as the first step for high-level applications. Pixel based approaches usually ignore the spatial coherence, while region based approaches are sensitive to region size and scene complexity. In this paper, we propose a robust background subtraction approach via multiple features based shared models. Each shared model is represented by a sequence of samples based on sample consensus. Each pixel dynamically searches a matched model around the neighborhood. This shared mechanism not only enhances the robustness for background noise and jitter but also significantly reduces the number of models and samples for each model. Besides, we concatenate color and texture features as multiple features according to the discriminability and complementarity, so that each pixel can find a proper model more easily. Finally, the shared models are updated by random selecting a pixel matched the model with an adaptive update rate. Experiments on ChangeDetection benchmark 2014 show that the proposed approach outperforms the state-of-the-art methods.

In order to reduce the social panic triggered by plane accidents, an effective approach is urgently needed to look for the missing passenger planes and then comfort the family of person who lost their lives in the accident. By analyzing the inherent features and motion law of the accident plane, this paper puts forward a differential dynamic model based method for locating the plane, which can effectively compensate the inherent deficiencies of radio position finding method, such as invalid locating. Firstly, this paper analyzes the instant conditions of the plane at its crashing moment and how these conditions are affected by the gravity and airflow resistance during its unpowered falling process, and as well as the plane falling trajectory. Then we establishes a dynamic model to determine the plane falling position and its corresponding search & rescue region. Finally simulation study and sensitivity analysis demonstrate that our proposed approach is effective and has good robustness. This research is very significant to reduce the rescue cost of air crash as well as improve the success rate of rescue.

Cloud manufacturing transforms scattered resources and capabilities supplied by entity enterprise into integrated manufacturing cloud services, which can be realized through composing different services. A complete candidate service set is the foundation of developing a composite service with higher quality. In order to improve QoS (Quality of Service) of the composite service, the correlations between services are considered and then classified in the whole composition process. However, considering the correlation in composition process but ignoring it during matching process, such as the existing methods about service correlation, will affect QoS of the composite service. In this paper, an extended service modelling is proposed to describe service correlations, and then a reservation algorithm is designed to reserve services with correlation in service matching stage so as to make the composite service much better than before. Simulation results demonstrate the effectiveness of our model and algorithm.

With the upcoming data deluge of semantic data, the fast growth of ontology bases has brought significant challenges in performing efficient and scalable reasoning. Traditional centralized reasoning methods are not sufficient to process large ontologies. Distributed reasoning methods are thus required to improve the scalability and performance of inferences. This paper proposes an incremental and distributed inference method for large-scale ontologies by using MapReduce, which realizes high-performance reasoning and runtime searching, especially for incremental knowledge base. By constructing transfer inference forest and effective assertional triples, the storage is largely reduced and the reasoning process is simplified and accelerated. Finally, a prototype system is implemented on a Hadoop framework and the experimental results validate the usability and effectiveness of the proposed approach.

Analyzing time series data can reveal the temporal behavior of the underlying mechanism producing the data. Time series motifs, which are similar subsequences or frequently occurring patterns, have significant meanings for researchers especially in medical domain. With the fast growth of time series data, traditional methods for motif discovery are inefficient and not applicable to large-scale data. This work proposes an efficientMotifDiscovery method forLarge-scaletimeseries (MDLats). By computing standard motifs, MDLats eliminates a majority of redundant computation in the related arts and reuses existing information to the maximum. All the motif types and subsequences are generated for subsequent analysis and classification. Our system is implemented on a Hadoop platform and deployed in a hospital for clinical electrocardiography classification. The experiments on real-world healthcare data show that MDLats outperform the state-of-the-art methods even in large time series.

In the current era massive datasets in healthcare are becoming much more available for analysis, where numerous features are designed or constructed to represent a patient. Feature selection algorithms play a key role in reducing the data dimension thus speeding up the succeeding learning algorithms as well as improving predicting accuracy. How to select the appropriate subset of features with low redundancy is one of the interesting problem in feature selection. In this paper, we present a new feature selection algorithm which aims to select low redundant features in the setting of grouped variables. We adopt a global optimization method based on Lipschitz continuity and present evaluation results on several datasets, which demonstrates the correctness and effectiveness of our algorithm.

With the global trend of population aging in industrialized countries, efficient information and communication technologies (ICT) for aiding elders or patients' healthcare have attracted great research attentions. Among these technologies, health state sensing by wearable sensors and mobile phones is an important foundation. It monitors the real-time body states; stores, or sends the result to remote family members or doctors. In this way, it can either help people to pay more attention to the overlooked phenomenon, such as the clue of dangerous disease, or help people to issue panic alert when emergency happens. There are many critical issues in health sensing. First, the sensors must be non-intrusive to people's comfort and safety, while providing good accuracy. At the same time, because of being worn by people, numerous noises posed by body motions must be efficiently processed for reducing false alarming. At last, different health or decease signals generally require different sensing technologies and instrument. To tease out the technology advantages that address these challenges and diversities, this paper presented a survey on the state of the art of health sensing technologies using body sensor networks and mobile phones. It classify related works by their application goals, including i) fall detection, ii) gait analyzing, iii) activity qualification, iv) heart state sensing, and v) sleep sensing. It also conducts summary and comparison of related sensing systems and algorithms, to reveal the development lines in each subarea.

Recently, cloud computing has emerged as a new model of service provisioning, in this model, one of the most challenging problems is workflow scheduling, i.e., the problem of satisfying users' QoS while minimizing the execution cost of cloud workflow. This paper propose a workflow scheduling algorithm called Control Structure Reduction algorithm(CSR). In CSR, the workflows represented by DAG (Directed Acyclic Graph) can be converted into an equivalent sequence control structure by such means as mergers and reduction. Then we can easily identify the critical path of the workflow process. By using Time Float Distribution Algorithm, the total time float is allocated to each task based on critical tasks in critical path, eventually to enlarge the cost optimization intervals of all tasks. The simulation results show that CSR has a promising performance in decreasing the execution cost for large workflows.

Energy efficiency and sustainable development have been the focus of the world's attention. In order to promote the execution of energy reduction, energy control systems, which could operate the electrical appliances, are under research at present. Before putting the energy control systems into real buildings, comfort assessment and energy consumption analysis need to be conducted but such operations require a large number of test cases to ensure the stability and effectiveness of the systems. Nevertheless, real data collection from each building is tedious and expensive; and manual test data generation may drop some important effective factors or relationships. Therefore, a tool of test data generation, which could generate large volumes of test data, is desperately needed. In this paper, we propose a neural network model to generate a large test data set for comfort assessment and energy consumption analysis. This approach is based on an existing set of real-world data, and generalizes it into a larger data set. Our analysis indicates that the proposed approach is reliable and effective.

Cataract is one of the most common diseases that might cause blindness. Previous research shows that cataract occupies almost 50% in severe visually impairments. Considering the fact that retinal image is one of the most important medical references that help to diagnose the cataract, this paper proposes to use a neural network classifier for automatic cataract detection based on the classification of retinal images. The classifier building procedure includes three parts: preprocessing, feature extraction, and classifier construction. In the pre-processing part, an improved Top-bottom hat transformation is proposed to enhance the contrast between the foreground and the object, and a trilateral filter is used to decrease the noise in the image. According to the analysis of pre-processed image, the luminance and texture message of the image are extracted as classification features. The classifier is constructed by back propagation (BP) neural network which has two layers. Based on the clearness degree of the retinal image, the patients' cataracts are classified into normal, mild, medium or severe ones. The initial evaluation results illustrate the effectiveness of our proposed approach, which has great potential to improve diagnosis efficiency of the ophthalmologist and reduce the physical and economic burden of the patients and society.

Traditional collaborative filtering (CF) methods suffer from sparse or even cold-start problems, especially for new established recommenders. However, since there are now quite a few recommender systems already existing in good working order, their data should be valuable to the new-start recommenders. This paper proposes shared collaborative filtering approach to leverage the data from other parties (contributor party) to improve own (beneficiary party's) CF performance, and at the same time the privacy of other parties cannot be compromised. Item neighborhood list is chosen as the shared data from the contributor party with considering differential privacy. And an innovative algorithm called neighborhood boosting is proposed to make the beneficiary party leverage the shared data. MovieLens and Netflix data sets are considered as two parties to simulate and evaluate the proposed shared CF approach. The experiment results validate the positive effects of shared CF for increasing the recommendation accuracy of the beneficiary party. Especially when the beneficiary party's data is quite sparse, the performance can be increased by around 10%. The experiments also show that shared CF even outperforms the methods that incorporate the detailed original rating scores of the contributor party without considering the privacy issues. The proposed shared CF approach obtains a win-win situation for both performance and privacy.

Workflow model analysis is performed at logic, temporal, and performance levels. This paper mainly deals with the performance level issues. Workflow net (WF-net) is extended with time information to the timing workflow net (TWF-net). To provide a formal framework for modeling and analyzing workflow, this paper proposes a multidimension workflow net (MWF-net) that include multiple TWF-nets and the organization and resource information. The algorithm to decompose a free-choice and acyclic Petri nets (PN) into a set of T-components is extended to a TWF-net containing iteration structures. Then, resource availability and workload analysis is performed. A method for computing the lower bound of average turnaround time of transaction instances processed in a MWF-net is proposed. Finally a case study is used to show that the proposed method can be effectively utilized in practice.

Multi-dimension Workflow net (MWF-net), which includes process, organization, and resource perspectives, is introduced. Using the structure analysis of the TWF-net in the process perspective and the perspectives mapping, the routing of transaction instances in the multi-TWF-nets can be projected into the flow of transaction instance between different resource pools in the resource perspective. After the relevant work in is briefly reviewed, the boundedness verification method of a MWF-net is proposed. A MWF-net is bounded implies the corresponding queuing network in the resource perspective has stable solution. Based on the discussion of several operational principles in the context of workflow model, an approximate method for performance analysis of a workflow model is presented.

The analysis of the correctness and rationality of a workflow model plays an important role in the research of workflow techniques and successful implementation of workflow management. This paper points out the relevant problems in the verification and analysis of a workflow model. It discusses two important properties: schedulability and boundedness of a workflow model considering timing constraints. To specify the timing constraints, WorkFlow net is extended with time information, leading to timing constraint workflow net (TCWF-net). This paper presents a model mapping method to convert a directed network graph (DNG) based workflow model, which is built by a graphic process modeling language extended with time information, into a TCWF-net. It then discusses its schedulability verification and synthesis. An algorithm to decompose an acyclic and free-choice TCWF-net into a set of T-components is presented, followed by a boundedness verification method. The usefulness of the research results is illustrated by an example.