As the application of AI in cervical cancer cell detection expands, the demand for large-scale labeled pathological slide data has increased, resulting in a time-consuming and costly process. Furthermore, significant domain shifts caused by variations in sample collection, processing, and staining conditions across different medical institutions limit the generalization capability of detection models. To address these challenges, existing unsupervised domain adaptation (UDA) methods attempt to generalize source domain models to target domains with unannotated data, reducing reliance on annotated datasets. However, these approaches still face challenges, such as the unavailability of source data, label noise leading to model degradation, and insufficient convolutional feature extraction. To overcome these limitations, this paper proposes a Deformable-based Source-Free Unsupervised Domain Adaptation (DSFUDA) method for cervical cell detection, which utilizes only on the source domain model and unlabeled target domain data. The proposed approach introduces a novel source-free UDA framework for cervical cell detection, incorporating a denoising module based on a diffusion model (FD-Module) to suppress erroneous and domain-specific features, preventing model degradation. Additionally, a deformable context-aware module (DCA-Module) is developed to adaptively adjust convolutional receptive fields, enhancing feature extraction and improving the accuracy of abnormal cell detection. The effectiveness of the proposed method is validated on two public datasets, CDTBS and Sipakmed.

In the financial field, options hedging has gained significant attention due to its crucial role in risk management and corporate operations. Traditional option hedging methods, such as delta hedging based on the Black-Scholes model, face limitations in practical application due to assumptions of constant volatility and the neglect of transaction costs. In recent years, reinforcement learning (RL) has become a hot topic in option hedging research because it can interact with the environment and adjust hedging strategies based on environmental feedback, better reflecting real market conditions. However, it still faces several challenges, such as the neglect of historical market information, insufficient evaluation of hedging cost distributions, and high randomness in model training. This paper proposes a reinforcement learning method for option hedging using quantile regression and curriculum learning with historical data fusion, aiming to address the current shortcomings in historical information utilization, risk assessment of the hedging strategy, and robustness of the model. Firstly, historical market information is incorporated into state variables, and the time-trend multi-head self-attention mechanism (TiTrMHSA) and Gated Recurrent Units (GRU) are introduced to capture the dynamic changing trends of historical information and integrate current features, significantly improving the hedging strategy’s sensitivity to market fluctuations. Secondly, quantile regression is used in the value network, combining the Quantile Huber Loss function to fit the hedging cost distribution. This enables a comprehensive evaluation of strategy performance and enhances risk control capabilities. Finally, by incorporating the concept of curriculum learning, a two-stage training method for the agent is designed to progressively optimize the agent’s learning process from simple to complex, addressing the issue of high randomness in early-stage training. The Experimental results show that this method significantly outperforms traditional BS-Delta hedging and other reinforcement learning models regarding hedging costs and the balance between returns and risks, demonstrating superior robustness and adaptability.

As the demand for personalized assembly increases, effective assembly task planning continues to present significant challenges, particularly in scenarios with diverse assembly goals and complex dependencies among components. To address these challenges, we propose the Spatio-Temporal Scene Graph-Enhanced Assembly Planning Model (STG-AP), which utilizes only the initial and goal visual observations of the assembly task to infer the assembly sequence from a global perspective, leveraging Graph Neural Network (GNN) and Transformer architectures to capture spatiotemporal dependencies. To evaluate the model’s performance, we conducted assessments on the IKEA ASM Dataset and also developed a lightweight Block-Assembly (Block ASM) Dataset designed to offer a rich variety of scenarios for training and evaluation. Through extensive experiments, we demonstrate that STG-AP consistently outperforms state-of-the-art methods across various sequence lengths. Specifically, our proposed Spatio-Temporal Scene Graph (STG) effectively enhances the performance of long-sequence task planning by skillfully learning the latent representations of the scene. Our method provides a more robust and effective solution for long-sequence flexible assembly planning in robotics.

Extracting structured information and key details from medical report narratives is crucial to support healthcare data management, analysis and decision-making. However, the specialized nature of the reports, the complexity of the contents, and the high accuracy requirements of the results pose significant challenges to the structuring task. In this paper, we develop an LLM-based method to extract structure information from medical report narratives. Defining the structuring problem as mapping the narrative reports to the domain ontology, we design a framework to develop specialized LLMs that automatically learn and establish the mappings. At the core of this framework are report partitioning and interactive training data generation modules are. By separating complete reports into logically independent segments and training the LLMs on these segments independently, the trained LLMs can accurately capture the semantic relationships within each segment. Additionally, we explore different LLMs and formulate a simplistic scoring method to compare their accuracy, enabling us to select the best-performing model. Experimental evaluation on a real-world breast ultrasound report dataset demonstrates that our method achieves high accuracy with a small training dataset (400 samples). Specifically, the accuracy of structural information extraction and the attribute-value matching accuracy both exceed 96%.

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder, where early diagnosis is critical for slowing disease progression. Cognitive assessments, known for their high sensitivity and specificity, are essential tools for early detection. However, clinical data from AD patients are often incomplete, posing significant challenges in predicting cognitive scores and accurately assessing disease progression. This paper proposes a time-decay LSTM network combined with a Multi-Condition Variational Autoencoder (MC-VAE) to address these challenges. First, the MC-VAE with clustering is employed to effectively impute missing values, capturing the diversity of the data distribution while maintaining its discreteness during interpolation. Next, a multi-representation information encoder, composed of several MUST-Encoding components, encodes the imputed data. This process effectively captures the non-linear and non-periodic temporal patterns arising from irregular time series. Finally, the encoded information is input into a Time-Decay LSTM network. Experimental results validate that our model outperforms conventional methods in predicting AD progression for the ADNI and NACC public datasets.

Automatic tracking and segmentation of lesions in ultrasound videos could assist in early diagnosis and treatment plan development. However, this task is quite challenging due to problems such as low visual saliency of the lesions and large variation between adjacent frames. In this paper, we develop a Spatial-Temporal Fusion Network (STFNet) for unsupervised ultrasound video object segmentation. First, an Edge Blur Enhancement Module is designed to extract and preserve the edge details of the target objects in ultrasound frames for spatial feature enhancement. Then, a Dynamic Alignment Module is developed to correct the inter-frame inconsistencies by aligning target objects from adjacent frames with those in the current frame for temporal feature enhancement. To incorporate both the spatial and temporal information, we further implement a mixed training strategy. These innovations collectively refine the model’s learning process and substantially boost segmentation accuracy. Extensive evaluations on real lymphoma ultrasound video data demonstrate the competitive segmentation results of STFNet. Specifically, compared with the best results among seven competing baselines, STFNet achieves the best scores in terms of region similarity ${\mathcal{J}}$, contour accuracy ${\mathcal{F}}$ as well as their average ${\mathcal{J}}\& {\mathcal{F}}$.

Financial market volatility forecasting plays an extremely important role in the financial field. It helps investors, traders and financial institutions more accurately assess market risks and formulate effective investment strategies. However, due to the uncertainty and complexity of financial markets, the non-stationarity and volatility of financial time series make volatility forecasting more challenging. At the same time, when forecasting the volatility of multiple financial products, the distinct market behaviors and risk characteristics of different products may lead to distortion and overfitting when using the same forecasting model. Consequently, it is necessary to construct a separate volatility forecasting model for each financial product, which in turn brings about the problems of difficulty in obtaining data and insufficient samples. To address these issues, a financial market volatility forecasting method based on domain adaptation is proposed. The method first constructs a volatility forecasting model FinWaveNet for single-domain financial products. FinWaveNet introduces causal convolutional layers and inflated convolution to capture financial time series features more comprehensively. Residual networks optimize data feature capture, while Gated Recurrent Units (GRU) and highway networks ensure stable information transmission, addressing non-stationarity and volatility challenges and enhancing the accuracy in single-domain scenarios. Subsequently, we extend our approach to construct a financial volatility forecasting model using domain adaptation methods, which combines adversarial learning and maximum mean difference (MMD) method to address data feature misalignment across different domains, thereby improving the accuracy in the multi-domain scenario. Finally, through ablation and comparison experiments, the paper validates the effectiveness of the proposed method in financial field volatility forecasting, demonstrating optimal performance based on the MAE evaluation index.

Real-time tracking and segmentation of ultrasound video sequence are prerequisite for identifying and analyzing lesions. While significant progress has been made in natural video object segmentation, developing a model for ultrasound video is still challenging due to problems such as low distinguishability and low visual saliency of the target objects, large variation between adjacent frames. These challenges are inherently complex and cannot be effectively tackled through a single process. This paper develops an end-to-end multi-stage network (EMNet) for ultrasound video object segmentation. EMNet consists of two stages. The inital mask generation stage comprises a contrast-enhanced layer to enhance visual contrast between targets and backgrounds. In this stage, a module that adopts the encoder-attention-decoder structure is designed for mask induction. After obtaining the initial segmentation mask, the mask refinement stage is followed to further improve initial segmentation. To prevent the propagation of errors, a gating mechanism is designed to control the fusion of segmentation probability maps in the initial and refinement stages. By transforming certain fixed parameters in different stage into trainable parameters and establishing an end-to-end learning process, we optimized the performance of our approach. We evaluate EMNet on real-world lymphoma ultrasound video dataset. Compared with the best results among seven competing baselines, EMNet achieves the best performance in terms of ℐ&ℱ and ℱ measures, the second-best performance with Param and FPS measures, which demonstrates the competitive performance in terms of both speed and accuracy.

In the era of big data, data security poses a paramount challenge in dealing with large-scale data in cloud computing environments. Traditional encryption methods such as RSA and Paillier face the problems that, as the data volume increases, the encryption consumes too much computation resource and the storage size of ciphertext is too large. To tackle these problems, this paper explores secure and lightweight tensor train (TT) based encryption methods for large-scale data in federated cloud environments. First, we develop a novel encryption method called Orthogonalized TT Encryption (OTT-EPT), which can encrypt large-scale plaintext data into a list of small-size ciphertext (TT cores) and store them in different clouds. Second, to tackle the unique characteristics of TT decomposition that adjacent TT cores have the same rank, which could be exploited by attackers to threaten encrypted data, we further propose a Noise-Added Orthogonalized TT Encryption (NOTT-EPT) method. NOTT-EPT can change arbitrary TT-rank by adding noise TT-cores on specific order. Furthermore, we present a TT-based encryption framework and a suite of secure homomorphic operation protocols for large-scale data computation in federated cloud environments. The experimental results indicate that, compared to state-of-the-art methods, our proposed methods not only ensure the integrity and security of encrypted data but also reduce both time and space consumption by about 1000 times.

Convolutional neural networks have achieved great success in the field of image recognition, however, their inherent black-box properties hinder their application in highly sensitive fields such as healthcare and autonomous driving. Therefore, interpretable artificial intelligence (XAI) has become a hot topic of academic research, which aims at exploring the working mechanism inside deep convolutional neural networks. Causality is an important means to explore the working mechanism of neural networks. For image classification, the image classification results depend on the causal features or background features of the input image. Grad-CAM is a popular interpretation method to visualize these features, and its visualization results in a deep integration of causal and context features. In this paper, we propose a background category-based feature integration method to estimate the context features, and then apply an ensemble theory approach to extract causal features from Grad-CAM. The experimental results show that our Grad-CAM-based extracted causal features not only have better causality in ResNet50, VGG16, and GoogleNet compared with Grad-CAM and Grad-CAM++, but also are similar to the visualization results of Grad-CAM ++ in terms of target object localization of visualization results. In addition, we verify that our method is applicable not only to high category fine-grained datasets but also to low fine-grained datasets through category fine-grained experiments.

In recent years, with the rapid development of mass spectrometry, there are more and more cases of finding biomarkers by proteomics for clinical diagnosis and treatment, and artificial intelligence is gradually replacing traditional statistical learning methods, but the process of finding biomarkers is not yet uniform and perfect, and there are gaps in validation. With this aim, this study proposes a pipeline for the analysis of biomarkers based on data independent acquisition(MS-BioAP). MS-BioAP includes three modules: mass spectrometry protein identification module, differential protein analysis module, and classification model building module. Combining statistical learning and artificial intelligence methods to complete the biomarker search, we also propose a target association scoring algorithm(TAScore), use the large amount of prior knowledge in the knowledge graph to complete the biomarker validation, and create a classification model based on the selected biomarkers. We conducted experiments using a clinical stress response mass spectrometry dataset and the final selected biomarkers achieved an accuracy of 1.0 in the classification model with both high sensitivity and specificity. The analytical pipeline is complete and comprehensive, with clinical implications. In the future, we will continue to optimize the analytical pipeline and validate this analytical pipeline with additional datasets.

Mycophenolic acid (MPA) is a commonly used immunosuppressive drug.The anti-immune rejection effect of mycophenolic acid is closely related to its exposure level in the body.In clinical practice, mycophenolate acid drug exposure level is usually reflected by monitoring the area under the drug-time curve MPA-AUC0-12h.Calculating the MPA-AUC0-12h requires numerous blood sampling time points.Not only does the medical staff have more work, but patients suffer more as well.Limited sampling strategies (LSS) are generally used to reduce the number of time points.Nevertheless, this method involves complicated calculations and the predictive accuracy is very low for small sample data.A new method of predicting the MPA-AUC0-12h value is proposed based on the selection of SHAP features with an improved neural network for small sample data.The experimental results show that the average prediction errors of the MPA-AUC0-12h values on different data sets by our method are better than that of the baseline models.

Breast cancer has become one of the most common malignant tumors in women worldwide, and it seriously threatens women’s physical and mental health. In recent years, with the development of Artificial Intelligence(AI) and the accumulation of medical data, AI has begun to be deeply integrated with mammography, MRI, ultrasound, etc. to assist physicians in disease diagnosis. However, the existing breast cancer diagnosis model based on Computer Vision(CV) is greatly affected by the image quality; on the other hand, the breast cancer diagnosis model based on Natural Language Processing(NLP) cannot effectively extract the semantic information of the mammography report. The lack of model interpretability also makes the existing diagnostic models have low confidence. In this paper, we proposed Breast Cancer Causal XAI Diagnostic Model(BCCXDM). Specifically, we first structured the mammography report. Then find the causal graph based on the structured table. We combine the existing tabular learning method TabNet with causal graphs(Causal-TabNet) to enable reasoning in the graphs to preserve the correlation between features. More importantly, we use GNN and node transition probability to aggregate node information. We evaluate our model on the real-world mammography report, and compare it with other popular interpretable methods. The experimental results show that our interpretable results are closer to the diagnostic criteria of clinicians.

Pneumonia kills more than 1.4 million children per year. It is among the top diseases which caused many deaths around the world. Because the diagnosis by chest x-ray films is rather complicated, the accuracy of diagnosing pneumonia is not high even by expert radiologists. In this study, the implementation of a deep learning technique for chest x-rays is presented. For this, we have proposed a new method to simplify the process of pneumonia detection. A convolutional neural network (CNN) model “ConvNet-21” has been proposed to extract important features from the chest x-ray images. The dataset is divided into two categories: training and testing, with each category further subdivided into normal and pneumonia. Several data augmentation techniques are applied to the dataset. In this study, five different models have been discussed. There are five convolutional layers in the proposed CNN model and the remaining four are transfer learning models, VGG16, VGG19, Inception-V3, and ResNetl52V2. We have analyzed the performance of all these models by using different performance measure parameters. The accuracy obtained by our proposed CNN model is SS.94%. The accuracy obtained by VGG16, VGG19, Inception-V3 and ResNetl52V2 are 92.7S%, 91.50%, 74.19% and 77.72%, respectively.

Alzheimer’s Disease (AD) is an irreversible neurodegenerative disease that commonly occurs in the elderly. With the current accelerated aging process, the accurate diagnosis of early AD is essential for patient care and disease delay. In recent years, Magnetic Resonance Imaging (MRI) has become increasingly important in diagnosing AD due to advances in deep learning and neuroimaging technology. This paper proposes a model framework for multi-classification prediction of Alzheimer’s disease based on fusing multi-modal features. Firstly, the sMRI data are pre-processed based on ROI templates with different segmentation accuracy levels to extract morphological features including gray matter volume, surface area and cortical thickness, and then these features are combined with the corresponding Clinical Data to produce the Indicators dataset. Secondly, a 3DCNN-SE module is proposed to extract the primary features from the 3D MRI data. In order to reduce the dimensionality of the Indicators, an Indicator Selection Strategy is designed to select the most relevant features from the Indicators. Finally, a Multi-Attention-Fusion Module (MAFM) is developed to perform multi-modal data fusion on the results of feature extraction and selection, followed by a SoftMax classifier for AD disease diagnosis. We evaluated 596 patients from the Alzheimer’s Disease Neuroimaging Initiative(ADNI), including 198 patients with Alzheimer’s disease (AD), 200 patients with mild cognitive impairment (MCI), and 198 patients cognitively normal(CN). As a result, 88% accuracy is achieved on the three classifications, which is better than the related methods mentioned in literature.

Lung diseases are among the most widely recognized reasons for serious disease and passing around the world. The timely analysis is crucial to lessen the risk of any disease and if any disease is diagnosed, then precautions or medicines are immediately given. Therefore, for the diagnosis of lung sound auscultation, progressive computational tools are established and play a very vital role in the detection of disease-related anomalies. The aim of this study is the joint learning of the model that only extracts important breathing samples without generating redundant noise, and then uses this information to train lung sounds into four categories: normal, wheezes, crackles, and wheezes and crackles. This paper signified an unsupervised approach that depends on a Denoising Auto-Encoder (DAE). It utilizes the rebuild errors between the input and the output of the Auto-encoder as an activation audio signal to identify noisiness. A novel design of Recurrent Neural Network (RNN) called noise-masking anomalies recurrent neural network (NMA-RNN) for lung sound order is projected. ICBHI database used in this paper and some results of previous models were compared and achieved 95% accuracy.

Summary Entity recognition plays an important role in building the electronic medical records (EMRs) based medical knowledge graph, which is significant for building Clinical decision support (CDS) system. Cross‐disease clinical documents are context‐related and have different interrelated semantic structures, which bring challenges for entity recognition using traditional methods. In order to solve these problems, this paper proposes a co‐training based entity recognition approach for cross‐disease clinical documents. In this model, we first build partial annotation corpus of the single disease using dependency syntax analysis and the medical statement rule unifies. Then, according to the partial annotation corpus of different diseases, the sentence level features are extracted through the Bi‐LSTM layer with memory unit and CRF methods, which optimize the whole sequence and improve the combination probability of sequence labels. Finally, the results with higher confidence are selected by cross feedback to label the corpus, which enlarges the size of corpus and improves the accuracy of the document entity recognition. The experiment result proves the availability and high efficiency of our method.

With the development of information technology, most of the medical institutions have established the web medical platform, and it will have a large number of patients' evaluating textual information. These subjective text messages contain emotional information, such as views, opinions and attitudes of patients. It takes a lot of manpower and time to analyze and evaluate positive and negative evaluations by manual methods. Therefore, this paper presents a method of emotion evaluation of medical reviews based on character-level vector convolution neural network. Aiming at the problem of text input noise caused by word segmentation and polysemy, ignoring the structure information of sentence in traditional convolution neural network model, this paper proposes a segmentation pooling convolution neural network model based on character-level vector. Using the improved skip-gram model to train the character vector, and using different convolution kernels of different sizes to extract the sentence features, this proposed model then use the method of segmentation pooling to preserve the maximum eigenvalues of each part ofthe sentence. The experiments show that the accuracy of the proposed model in the emotional analysis of medical texts is about 12% higher than that ofthe traditional convolutional neural network model. In the actual task of emotion analysis of medical texts, the accuracy of the model is as high as 88.2 %.

The deployment of wireless sensor networks (WSNs) is expected to have a significant impact on the efficiency of many civil and military applications, such as disaster management, environment monitoring, combat field surveillance and space exploration. In unattended WSN setups, most of the energy aware routing approaches pursue multi-hop paths in order to minimize the total transmission power. However, the hops close to the base-station (also known as the sink) become heavily involved in packets receiving and forwarding and thus their batteries get depleted rather quickly, and sometimes become traffic bottlenecks. The failure of these nodes also can create a void around the sink and significantly increase the energy consumed in communication with the sink. This paper evaluates the effectiveness of three approaches for handling the weakened network topology caused by the failure of sensors around the sink. The approaches are: (1) increasing the population of sensors in selected areas; (2) increasing the number of sink nodes in the network; and (3) repositioning the existing sink. The performance of the three approaches is validated in a simulated environment.