Clustering techniques offer strong interpretability. However, they have significant limitations in the deep learning area due to their difficulty in capturing complex data structures, such as spatial and contextual information. This issue is especially pronounced in neuroimaging research, where high-dimensional and complex data greatly restricts the feature representation capability of clustering models. Hence, we propose a Neuroimaging Deep Clustering (NIDC) backbone network. We convert 3D neuroimagings into point sets and design clustering-based paradigms for context feature aggregation, feature interaction, and feature dispatching to enable deep feature extraction from the point sets. To better capture spatial information, we propose brain spatial relative position encoding, which assists the clustering paradigm in better understanding the anatomical structure of brain tissue and the positional relationships between different regions. Additionally, we design a sample center loss function to encourage tighter clustering of labels or voxels/feature points of the same class in the feature space, aiming to suppress both inter-class and intra-class similarities. Meanwhile, NIDC preserves the interpretability of traditional clustering techniques, allowing it to uncover relationships between brain regions and trace the decision-making process of each voxel. NIDC achieves highly competitive performance across multiple datasets in various downstream tasks, emerging as a new and practical solution for neuroimaging analysis. Code is available athttps://github.com/IMCTGD/NIDC.

In the diagnosis of neurodegenerative diseases, effectively integrating heterogeneous data from medical imaging and clinical diagnostic texts remains a significant challenge. In this study, we propose CAFNet, a novel collaborative learning frame-work designed to bridge the gap between structural MRI and clinical narratives through anatomically guided and symptom-aware fusion. The framework leverages a large language model to encode rich semantic information from diagnostic text, while a anatomical MaxPooling Convolution module extracts region-specific features from brain MRI volumes using directional pooling along canonical axes. These heterogeneous features are further aligned via a symptoms attention convolution and refined through a Category-Aware Filtering module based on mutual information, which adaptively enhances class-relevant repre-sentations. Experimental evaluations on three public datasets demonstrate that CAFNet significantly improves diagnostic accuracy and generalization, outperforming existing state-of-the-art multimodal classification models. The results confirm the effectiveness of integrating anatomical priors and clinical semantics for interpretable and robust brain disease classification.

One-shot talking head synthesis aims to animate a source person’s portrait with driving video sequences. Recent facial keypoint-based methods have achieved remarkable animation performance and produced high-quality results. However, it remains challenging to perform cross-identity face reenactment by transferring subtle facial motions with correct geometry and appearance. To break the above limitations, in this paper, we propose a subtle motion compensation network to recover correct facial expressions by leveraging the decoupled 3D Morphable Model (3DMM) coefficient. In addition, to generate faithful animation results, a facial appearance feature memory bank is designed to learn accurate facial features and better recover the appearance. Experimental results have demonstrated that our proposed model can outperform state-of-the-art methods by generating faithful videos with correct subtle motion transfer and consistent identity preserving.

In recent years, deep reinforcement learning (DRL) has demonstrated significant potential to address complex and dynamic supply chain optimization (SCO) problems. However, existing DRL algorithms often encounter challenges when dealing with large-scale and high-dimensional supply chain decisions, making it difficult to effectively coordinate production and transportation. To address these issues, this paper proposes an innovative deep reinforcement learning algorithm—Action Decomposition Actor-Critic (ADAC). This algorithm significantly reduces learning complexity by decomposing complex decision tasks into multiple subtasks. Based on real-world supply chain scenarios, we construct a complex multi-stage supply chain environment. Additionally, we employ action decomposition-based Proximal Policy Optimization (PPO) and Soft Actor-Critic (SAC) algorithms to learn optimal policies in continuous action spaces, enabling fine-grained control of production and transportation. To verify the effectiveness of the algorithm, we conduct extensive experiments on both simulation and real-world datasets. Experimental results demonstrate that the ADAC algorithm outperforms traditional heuristic algorithms and general DRL algorithms in multiple complex supply chain scenarios. This shows the strong robustness and wide applicability of the ADAC algorithm in SCO problems.

Alzheimer’s disease (AD) is the leading cause of dementia in the elderly and its incidence is rapidly increasing. sMRI data is important for early diagnosis of AD, but the scarcity of its labelled data makes supervised learning not good enough to train powerful deep learning models. Self-supervised learning (SSL) is a promising deep learning method that allows models to automatically learn their own features through a pretext task, and has achieved great success in natural image analysis. We propose AD-MAE for AD early classification with a self-supervised model based on a mask autoencoder (MAE) that employs a masking strategy tailored for sMRI. In order to capture both local details and global features, we incorporate a contrast learning (CL) branch, while employing a fusion loss function to address false positives and negatives in classification. Experimental results show that AD-MAE performs well in different task settings on the ADNI dataset and is expected to provide strong support for early diagnosis of AD.

The integration of multi-modal data and deep learning techniques is one of the important research directions for the automatic diagnosis of Alzheimer’s disease (AD). Despite the promising performance of existing deeplearning-based multi-modal neuroimaging models, these models often either fail to distinguish and utilize the unique strengths of FDG-PET in highlighting metabolic activity and the high-resolution depiction of anatomical details by sMRI, or neglect the complementary relationships between the two modalities. To address the above limitations, we propose an adaptive mask attention dual branch CNN-Transformer model. First, the dual-branch feature extractor that simultaneously captures the unique pathological changes in both sMRI and FDG-PET data while identifying common patterns between the two modalities. Next, adaptive masked attention facilitates interaction between different modalities and identifies common features between them. Additionally, a fusion module performs adaptive fusion of multi-modal features at both the channel and spatial levels. Ultimately, the Kolmogorov-Arnold Network (KAN) is utilized in place of the traditional Multi-Layer Perceptron (MLP) for AD decision-making. The experimental results on the ADNI database show that the proposed model achieves excellent performance in different tasks, surpassing existing multi-modal AD prediction models.

Pre-extracted aortic dissection (AD) centerline is very useful for quantitative diagnosis and treatment of AD disease. However, centerline extraction is challenging because (i) the lumen of AD is very narrow and irregular, yielding failure in feature extraction and interrupted topology; and (ii) the acute nature of AD requires a quick algorithm, however, AD scans usually contain thousands of slices, centerline extraction is very time-consuming. In this paper, a fast AD centerline extraction algorithm, which is based on a local conformal deep reinforced agent and dynamic tracking framework, is presented. The potential dependence of adjacent center points is utilized to form the novel 2.5D state and locally constrains the shape of the centerline, which improves overlap ratio and accuracy of the tracked path. Moreover, we dynamically modify the width and direction of the detection window to focus on vessel-relevant regions and improve the ability in tracking small vessels. On a public AD dataset that involves 100 CTA scans, the proposed method obtains average overlap of 97.23% and mean distance error of 1.28 voxels, which outperforms four state-of-the-art AD centerline extraction methods. The proposed algorithm is very fast with average processing time of 9.54s, indicating that this method is very suitable for clinical practice.

Adaptive interpretable ensemble model based on three-dimensional Convolutional Neural Network (3DCNN) and Genetic Algorithm (GA), i.e., 3DCNN+EL+GA, was proposed to differentiate the subjects with Alzheimer's Disease (AD) or Mild Cognitive Impairment (MCI) and further identify the discriminative brain regions significantly contributing to the classifications in a data-driven way. Plus, the discriminative brain sub-regions at a voxel level were further located in these achieved brain regions, with a gradient-based attribution method designed for CNN. Besides disclosing the discriminative brain sub-regions, the testing results on the datasets from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and the Open Access Series of Imaging Studies (OASIS) indicated that 3DCNN+EL+GA outperformed other state-of-the-art deep learning algorithms and that the achieved discriminative brain regions (e.g., the rostral hippocampus, caudal hippocampus, and medial amygdala) were linked to emotion, memory, language, and other essential brain functions impaired early in the AD process. Future research is needed to examine the generalizability of the proposed method and ideas to discern discriminative brain regions for other brain disorders, such as severe depression, schizophrenia, autism, and cerebrovascular diseases, using neuroimaging.

The integration of structural magnetic resonance imaging (sMRI) and deep learning techniques is one of the important research directions for the automatic diagnosis of Alzheimer's disease (AD). Despite the satisfactory performance achieved by existing voxel-based models based on convolutional neural networks (CNNs), such models only handle AD-related brain atrophy at a single spatial scale and lack spatial localization of abnormal brain regions based on model interpretability. To address the above limitations, we propose a traceable interpretability model for AD recognition based on multi-patch attention (MAD-Former). MAD-Former consists of two parts: recognition and interpretability. In the recognition part, we design a 3D brain feature extraction network to extract local features, followed by constructing a dual-branch attention structure with different patch sizes to achieve global feature extraction, forming a multi-scale spatial feature extraction framework. Meanwhile, we propose an important attention similarity position loss function to assist in model decision-making. The interpretability part proposes a traceable method that can obtain a 3D ROI space through attention-based selection and receptive field tracing. This space encompasses key brain tissues that influence model decisions. Experimental results reveal the significant role of brain tissues such as the Fusiform Gyrus (FuG) in AD recognition. MAD-Former achieves outstanding performance in different tasks on ADNI and OASIS datasets, demonstrating reliable model interpretability.

Time series data with missing values are ubiquitous in real applications due to various unforeseen faults during data generation, storage, and transmission. Time-Series Data Imputation (TSDI) is thus crucial to many temporal data analysis tasks. However, existing works usually consider only one of the following two issues: (1) intra-feature temporal dependency, and (2) inter-feature correlation, leading to the overlook of complex coupling information in imputation. To achieve more accurate TDSI, we design a novel imputation model called TABiG, which delicately preserves the short-term, long-term, and inter-feature dependencies by attention mechanisms in a delay error-reduced bi-directional architecture. That is, it leverages GRU to model short-term temporal dependencies and adopts self-attention mechanisms hierarchically to capture long-term temporal dependencies and inter-feature correlations. The multiple self-attention mechanisms are nested in a bi-directional structure to alleviate the problem of delay errors in RNN-like structures. To facilitate model training with higher generalization, a masking strategy that mimics various extreme real missing situations beyond the simple random ones has been adopted for generating self-supervised learning tasks. Comprehensive experiments demonstrate that TABiG significantly outperforms most state-of-the-art imputation counterparts. Complementary results and source code can be accessed at https://github.com/Zhang2112105189/TABiG

Alzheimer’s disease (AD) is a typical neurodegenerative disease involving multiple pathogenic factors. Early detection is the key to effective treatment of AD. However, most methods are developed based on data from a single modality, and ignore the relationships among subjects. In machine learning problems, hypergraph can be used to express the relationships between objects. In light of this, a framework for early diagnosis of Alzheimer’s disease based on multimodal hypergraph attention network is proposed in this paper. Specifically, we combine multimodal features to construct cross modal hypergraph, which represents the high-order structural relationships among subjects. Finally, a hypergraph attention network is used to fuse hypergraphs and perform the final classification. Our experimental results on the Alzheimer Disease Neuroimaging Initiative (ADNI) database show that our proposed method has better classification performance than the most advanced methods.

Anomalous left coronary artery from pulmonary artery (ALCAPA) is a serious cardiac anomaly, and surgical repair is the main treatment for ALCAPA patients in clinical practice. Recently, 3D printing has been widely adopted in the surgical planning of ALCAPA, which can give surgeons an intuitive structure of the heart especially the coronary arteries. However, before 3D printing is conducted, experienced radiologists need to manually segment the coronary arteries on computed tomography angiography (CTA) images, which is time-consuming, tedious and biased. On the other hand, automatic coronary artery segmentation with normal structures has been extensively studied in the community, but cannot be effectively applied to ALCAPA due to the significant variation of coronary artery structure in ALCAPA. In this paper, we propose ImageALCAPA, the first 3D CTA image dataset of ALCAPA. The proposed dataset contains 30 ALCAPA CTA images, which is of decent size compared with existing medical imaging datasets. We further propose a baseline method that performs multi-task 2D- 3D ensemble for automatic segmentation of ALCAPA. It is shown by experiment that our baseline method outperforms popular existing works on coronary artery segmentation. However, as the highest average Dice Similarity Coefficient of coronary arteries is merely 65%, there is still much room for improvement. To facilitate further research on this challenging problem, our dataset and codes are released to the public [1].

Based on some similarities between the knowledge system composed of the rules with priority orders and the sequential learning ahead masking (SLAM) model, an approach to enhance the generalization capabilities of the former with the help of the later is advocated. Firstly, the mapping from a rule to weights is realized. Secondly, the SLAM model is initialized to contain the knowledge from the rules and the generalization capabilities of the model are improved through the adjustment of the weights. Thirdly, based on the model, the approach can realize the incremental learning to grasp the knowledge containing in the newly added instances. Finally, the experimentations testify the obtained model has stronger generalization capabilities.

By continuing the control surfaces of the time-varying modifying factor analytic expression method, we construct a new algorithm to improve the properties of the fuzzy controller obviously. The algorithm has two merits: (a) Static error is erased. (b) Its program is very simple. Therefore, its operation speed is fast. The excellent quality control and rapid response real time system can be realized easily only by cheap general digital chip-computer. The paper presents theoretic proof, Simulation results illustrate the effectiveness of the approach.

The attribute reduction and rule generation (the attribute value reduction) are two main processes for knowledge acquisition. A self-optimizing approach based on a difference comparison table for knowledge acquisition aimed at the above processes was proposed. In the attribute reduction process, the conventional logic computation was transferred to a matrix computation along with some added thoughts on the evolution computation used to construct the self-adaptive optimizing algorithm. In addition, some sub-algorithms and proofs were presented in detail. In the rule generation process, the orderly attribute value reduction algorithm (OAVRA), which simplified the complexity of rule knowledge, was presented. The approach provided an effective and efficient method for knowledge acquisition that was supported by the experimentation.

The paper proposes a novel simulated annealing algorithm, which can achieve a satisfactory or optimizing result rapidly. The partly continuous algorithm (PCA) is optimized by the algorithm. The self-adaptive fuzzy controller is made of the two methods to control sophistic plants. The simulation results illustrate that the self-optimizing controller has fine performances to time-varied nonlinear plants of which mathematics models aren't known.

The paper outlines a self-optimizing approach for knowledge acquisition with adaptively incremental sampling, which fused the self-optimizing approach for knowledge acquisition and sampling approaches in order to improve the efficiency of knowledge acquisition effectively. The proposed sampling approach enabled us to dynamically and adaptively adjust the sample size according to the data mining algorithm's performance on the training samples so as to utilize the sample size as small as possible without reducing the accuracy of the knowledge model. Finally, the self-optimizing approach for knowledge acquisition with adaptively incremental sampling was applied to rule generation from the diagnostic decision table for rheumatoid arthritis in Chinese medical science. Experimentation results showed that the approach was much better than other algorithms both in efficiency and in accuracy.

This paper aims to provide a complete theoretical analysis for the creative methods. First, the feature set is used to express the creative solution, and some of the basic operators induced from different creative methods are defined. Next, the paper employs the information entropy theory to prove the usability of the creative method to produce creative solutions, and presents a detailed proof of the creating ability of the methods in an evolving framework. The prime contribution here is that the paper provides not only the theoretical foundation for the creative methods, but also the way and theoretical foundation for implementation of the system of creative design. Finally, an empirical study demonstrates that solutions, which meet the proposed theory, tend to score high on creativity evaluation by field experts.

Attribute reduction and rule generation (attribute value reduction) are two of the main processes of knowledge acquisition. A self-optimizing approach based on a difference comparison table for knowledge acquisition for these processes is proposed. For the attribute reduction process, conventional logic computation was replaced by matrix computation with some added concepts from evolutionary computation and used to construct the self-adaptive optimizing algorithm. In addition, some sub-algorithms and proofs are presented in detail. For the rule generation process, a value orderly reduction algorithm, which simplifies the complexity of rule knowledge, is presented. The approach provides an effective and efficient method for knowledge acquisition, which is supported by the experimentation.

In this paper an evolution model for creative design involving a mathematical model, evolution framework and implementation algorithms etc., is presented to effectively support the automation of creative design. The model takes the creative design process as an evolving process of a product prototype to the optimal scheme in accordance with a kind of measurement. This evolving process is built on the open genetic algorithm so as not to be confined by the initial population, which can get new individuals beyond the initial population by using creative reminding algorithms. At the same time, the model not only defines new operators based on the semantics of invention such as crossover, mutation, fitness function etc., but also provides an interface for integrating some other new methods of creative design as well as creative thinking patterns into the system. This model also supports a human-machine interface for receiving the guide from the design engineer online. Results of experiments with our implemented system, named AIE1.0, show that the model is promising.