Video-based dynamic facial emotion recognition (FER) is a challenging task, as one must capture and distinguish tiny facial movements representing emotional changes while ignoring the facial differences of different objects. Recent state-of-the-art studies have usually adopted more complex methods to solve this task, such as large-scale deep learning models or multimodal analysis with reference to multiple sub-models. According to the characteristics of the FER task and the shortcomings of existing methods, in this paper we propose a lightweight method and design three attention modules that can be flexibly inserted into the backbone network. The key information for the three dimensions of space, channel, and time is extracted by means of convolution layer, pooling layer, multi-layer perception (MLP), and other approaches, and attention weights are generated. By sharing parameters at the same level, the three modules do not add too many network parameters while enhancing the focus on specific areas of the face, effective feature information of static images, and key frames. The experimental results on CK+ and eNTERFACE’05 datasets show that this method can achieve higher accuracy.

Cardiovascular diseases pose a long-term risk to human health. This study focuses on the rich-spectrum mechanical vibrations generated during cardiac activity. By combining Fourier series theory, we propose a multi-frequency vibration model for the heart, decomposing cardiac vibration into frequency bands and establishing a systematic interpretation for detecting multi-frequency cardiac vibrations. Based on this, we develop a small multi-frequency vibration sensor module based on flexible polyvinylidene fluoride (PVDF) films, which is capable of synchronously collecting ultra-low-frequency seismocardiography (ULF-SCG), seismocardiography (SCG), and phonocardiography (PCG) signals with high sensitivity. Comparative experiments validate the sensor’s performance and we further develop an algorithm framework for feature extraction based on 1D-CNN models, achieving continuous recognition of multiple vibration features. Testing shows that the recognition coefficient of determination ($R^2$), mean absolute error (MAE), and root mean square error (RMSE) of the 8 features are 0.95, 2.18 ms, and 4.89 ms, respectively, with an average prediction speed of 60.18 us/point, meeting the re-quirements for online monitoring while ensuring accuracy in extracting multiple feature points. Finally, integrating the vibration model, sensor, and feature extraction algorithm, we propose a dynamic monitoring system for multi-frequency cardiac vibration, which can be applied to portable monitoring devices for daily dynamic cardiac monitoring, providing a new approach for the early diagnosis and prevention of cardiovascular diseases.

Three-dimensional face analysis is now a hot research topic in computer vision. The acquisition of 3D faces requires acquiring multi-view face and obtaining them by point cloud registration. In this paper, a deep learning method is introduced in the multi-view RGBD depth sensor system to acquire matching point pairs of image data. Thus, 3D spatial point pairs are obtained. Then, the initial poses between the point clouds of multi-view faces are obtained by rigid body transformation estimation, and the coarse registration of the point clouds is completed. On this basis, for fine registration of multi-view 3D face point clouds using the Iterative Closest Point (ICP) algorithm, this paper proposes a method to convert partially overlapping point cloud registration into sub-region point cloud registration through facial region positioning. In this letter, the maximum value of RMSE for the experimental results of point cloud coarse registration is 2.26 mm with the mean and standard deviation of 1.18 ± 0.33 mm; the maximum value of RMSE for the experimental results of point cloud fine registration is 1.47 mm with the mean and standard deviation of 0.91 ± 0.14 mm. It is confirmed that the proposed method is very robust and effective for the registration of multi-view 3D face point clouds.

Tongue diagnosis is a crucial method in traditional Chinese medicine (TCM) for obtaining information about a patient’s health condition. In this study, we propose a tongue image segmentation method based on deep learning and a pixel-level tongue color classification method utilizing machine learning techniques such as support vector machine (SVM) and ridge regression. These two approaches together form a comprehensive framework that spans from tongue image acquisition to segmentation and analysis. This framework provides an objective and visualized representation of pixel-wise classification and proportion distribution within tongue images, effectively assisting TCM practitioners in diagnosing tongue conditions. It mitigates the reliance on subjective observations in traditional tongue diagnosis, reducing human bias and enhancing the objectivity of TCM diagnosis. The proposed framework consists of three main components: tongue image segmentation, pixel-wise classification, and tongue color classification. In the segmentation stage, we integrate the Segment Anything Model (SAM) into the overall segmentation network. This approach not only achieves an intersection over union (IoU) score above 0.95 across three tongue image datasets but also significantly reduces the labor-intensive annotation process required for training traditional segmentation models while improving the generalization capability of the segmentation model. For pixel-wise classification, we propose a lightweight pixel classification model based on SVM, achieving a classification accuracy of 92%. In the tongue color classification stage, we introduce a ridge regression model that classifies tongue color based on the proportion of different pixel categories. Using this method, the classification accuracy reaches 91.80%. The proposed approach enables accurate and efficient tongue image segmentation, provides an intuitive visualization of tongue color distribution, and objectively analyzes and quantifies the proportion of different tongue color categories. In the future, this framework holds potential for validation and optimization in clinical practice.

This study investigates the feasibility of using traditional Chinese medicine (TCM) tongue features for diagnosing coronary heart disease (CHD). The aim is to propose an improved prediction method for CHD diagnosis. We collected data from 255 CHD patients and 308 non-CHD patients, including basic information and tongue image data. Using nine classic classifiers such as XGBoost and Random Forest, we evaluated their performance. To enhance prediction accuracy, we introduced an improved stacking model based on feature partitioning. This model partitions the data through a two-layer network, where the first layer extracts distribution features, and the second layer performs the final prediction. The results show that the proposed method achieved an F1 score of 85.20%, an AUC of 86.00%, a recall rate of 88.46%, an accuracy of 85.80%, a precision of 82.10%, an LR + of 5.39, and an LR- of 0.14. These findings suggest that using tongue image data for CHD prediction is effective, and the proposed method could serve as a valuable tool for non-invasive early detection, warranting further research and validation.

In practical scenarios, multi-view data often contains missing entries caused by complicated data collection and transmission procedures, posing great challenges to clustering analysis. Existing incomplete multi-view clustering methods have two obvious limitations: (1) Imputation-based methods inevitably generate inaccurate information during data recovery; (2) Imputation-free methods struggle to balance cross-view consistency and complementarity. To tackle the above issues, this paper proposes a deep incomplete multi-view clustering approach based on subspace learning, which integrates latent feature extraction, K-nearest neighbor-based feature imputation, cross-view contrastive alignment, and attention-driven fusion within a unified deep learning framework. Specifically, we first leverage autoencoders to extract view-specific latent features. Inspired by the observation that different views of the same sample should share similar latent representations, we impute missing latent features by leveraging cross-view correlations, using both view-specific complete features and common complete features through the K-nearest neighbor strategy. Then, we construct a fully-connected neural network layer to learn the self-expression matrix for each view. To minimize the impact of imputation noise on learning the self-expression matrix for clustering, these self-expression matrices are enforced to align using contrastive learning to capture cross-view consistency, and then fused based on a channel attention mechanism for pursuing cross-view complementarity. Finally, a high-quality cross-view consensus self-expression matrix is obtained for final clustering segmentation. Comprehensive evaluations on six benchmark datasets under various missing rates demonstrate the effectiveness and superiority of our method compared with state-of-the-art methods.