Group-level emotion recognition (GER) has emerged as a critical research topic for identifying collective emotions in multiperson scenarios. Despite recent advancements, such as dual branch cross-attention (CA) mechanism, existing methods struggle to differentiate ambiguous emotion categories effectively. In addition, the limited size and diversity of GER datasets hinder further performance improvements. To address these challenges, this article introduces a novel approach, the multimodal spatiotemporal semisupervised transformer (MSST). First, we propose a multimodal spatiotemporal transformer to encode spatial features, capture temporal dynamics, and fuse information from three modalities effectively. Second, a semisupervised learning (SSL) strategy leverages unlabeled data, enhancing robustness against noise and outliers. Last, a two-stage classification strategy and consistency loss are introduced to improve the model’s ability to handle category ambiguity and ensure robust predictions for similar samples. Comprehensive experiments conducted on benchmark GER datasets demonstrate that MSST either considerably outperforms or achieves competitive performance compared to state-of-the-art methods, underscoring its effectiveness in advancing GER research and overcoming the limitations of existing approaches.

Due to the explosive growth of electronic education records with highly sensitive nature, the security and authenticity of records have become an urgent issue to be addressed for data sharing. Although the blockchain-assisted searchable attribute-based encryption scheme provides certain trustworthiness for education records sharing, it still suffers from the risk of single key leakage and heavy computational overheads. In this paper, we propose a secure and efficient distributed sharing scheme with attribute-based searchable encryption (SEDS). On the one hand, we design a distributed public key searchable encryption method, which collaboratively generates distributed keys by decentralized blockchain nodes, effectively reducing the risk of single key leakage. On the other hand, we adapt and extend a fast pairing attribute-based encryption method to reduce the computational burden. We evaluate the performance of SEDS through theoretical analysis and experimental verification. The results show that SEDS can not only resist keyword guessing attacks but also reduce encryption and decryption time by 51% and 52% compared to traditional attribute-based encryption schemes, achieving a more secure and efficient education records sharing.

Federated Learning (FL) is a privacy-preserving machine learning paradigm, enabling decentralized devices to collaboratively train models without sharing local data. Traditional FL approaches, however, rely on averaging parameters across clients with homogeneous models, which limits their applicability in scenarios where clients require heterogeneous models. In this paper, we propose FedISMH, a novel approach to address model heterogeneity in FL. Instead of directly applying knowledge distillation, FedISMH clusters clients based on the structural similarities of client models, where clients’ structural features can be extracted through either labeled or unlabeled dataset This allows the proposed model to identify clients with similar model architectures while preserving privacy. Additionally, FedISMH introduces a dynamic mechanism to manage noise clients by aligning them with the most structurally similar clusters, ensuring that their inclusion promote the performance of the cluster. Experimental results on MNIST and SVHN demonstrate that FedISMH consistently outperforms state-of-the-art methods in both IID and Non-IID settings, offering improved accuracy, robustness, and flexibility in heterogeneous FL environments.

The automatic diagnosis of heart sounds is particularly important for cardiologists. However, the existing diagnostic methods still have a large space to be improved, In this paper, we proposed a novel method for heart sound classification. Our method consists of two stages. In the first stage, we preprocessed the heart sound signal, including two steps of denoising and downsampling, to reduce the noise and decrease the complexity of processing. In the second stage, we classify the processed signal, including framing and input network, and finally output three types of results. Our method was validated on the CirCor DigiScope Phonocardiogram Dataset. The result shows the F1 score reached 0.922 and is better compared to other networks’ results.

Screening vascular access dysfunction in hemodialysis requires tools that are objective and efficient. Listening for bruits during a physical exam is a subjective examination that can detect stenosis also called vascular narrowing when properly performed. Phonoangiograms (PAGs) which is a mathematical analysis of bruits increase the objectivity and sensitivity and permit quantification of stenosis. In this paper, we proposed Vision Transformer (ViT) for PAGs to automatically classify vascular stenosis. In particular, we added an auxiliary waveform to improve the classification performance. Moreover, we used the method of inter-patient verification to verify the performance of the proposed method. The experimental results show that the total F1 score, recall, and precision of the proposed method are 0.984, 1.000, and 0.969 respectively. We believe the method proposed in this paper has the potential to provide a reference for subsequent research.

ECG is an important medium for doctors to observe the working state of the patient’s heart, and the monitoring of patients based on ECG is very important in clinical diagnosis. The movement of the patient or the activities of other physiological organs in the process of collecting ECG will bring a lot of noise to the ECG signal acquisition, so it is necessary to de-noise the ECG signal with noise. In this paper, an U-net network based on Style-based Recalibration Module (SRM) channel attention is proposed to automatically denoise the noisy ECG signal. The ECG signals used are from clinical data marked by expert diagnostics. In addition, we also compared several popular denoising methods proposed in the past to conduct comparative experiments to verify the denoising performance of the model. The experimental results show that the proposed U-net network based on SRM channel attention can remove the noise contained in the ECG signal while retaining the characteristic shape of the ECG signal, and has a good effect in terms of signal-to-noise ratio and root mean square error.