Biomedical Signal Processing and Control最新文献

{"title":"Multi-level spatiotemporal graph attention fusion for multimodal depression detection","authors":"Yujie Yang,&nbsp;Wenbin Zheng","doi":"10.1016/j.bspc.2025.108123","DOIUrl":"10.1016/j.bspc.2025.108123","url":null,"abstract":"<div><div>Depression is a severe mental illness that affects hundreds of millions of people worldwide. In recent years, depression detection methods that integrate multimodal information have achieved significant results. However, limited by the small sample size of depression datasets, previous studies primarily focus on the impact of heterogeneous information in multimodal fusion, while deep interactions within each modality are often overlooked. Moreover, previous multimodal fusion methods often employed concatenation operations, which only allow modal features to be statically combined in the vector space and do not explicitly model the cross-modal semantic relationships. To address these issues, we propose a novel method named Multi-level Spatiotemporal Graph Attention Fusion (MSGAF), which enhances information interaction and sharing through multi-step fusion both within and between modalities. Specifically, within each modality containing multiple features, we designed a Multi-feature Temporal Fusion (MTF) module. The MTF module can fuse various features during the same time period to discover interactions among these features. For multimodal fusion, we adopt a multi-level fusion strategy to integrate these modalities, with the fusion process is represented as a Bidirectional Fusion Graph (BiFG). The graph attention mechanism is utilized to aggregate node information across the spatial neighborhood of the BiFG, which allows the graph structure to dynamically and adaptively capture the asymmetric relationships between modalities. Extensive experiments and analyses demonstrate the effectiveness of MSGAF, which achieves state-of-the-art performance on both the DAIC-WOZ and E-DAIC datasets. The code is available at: <span><span>https://github.com/wenbin-zheng/MSGAF</span><svg><path></path></svg></span></div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108123"},"PeriodicalIF":4.9,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144281022","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantitative identification of blood pressure information through directly decoding intrafascicular vagal activities","authors":"Mingcheng Gu ,&nbsp;Yiran Li ,&nbsp;Runhuan Li ,&nbsp;Xintong Wang ,&nbsp;Ting Xiang ,&nbsp;Yuan-Ting Zhang ,&nbsp;Cheng-Kung Cheng ,&nbsp;Jiguang Wang ,&nbsp;Xiaohong Sui","doi":"10.1016/j.bspc.2025.108140","DOIUrl":"10.1016/j.bspc.2025.108140","url":null,"abstract":"<div><div>The cervical vagus nerve is closely involved in blood pressure (BP) regulation, and the closed-loop vagus nerve stimulation (VNS) could mitigate concerns for neural adaptation by feeding back BP information to accomplish adaptive stimulation, making BP quantification significant. In this present study, BPs were directly decoded from vagal activities. The cervical left vagus nerves (LVNs) in eight normal Sprague-Dawley rats were intrafascicularly implanted with axon-like carbon nanotube yarn electrodes for vagal recording. Phenylephrine was intravenously injected to elevate the BPs, and LVN and systolic BP waveforms were synchronously recorded in real time. Then we quantitatively reconstructed beat-to-beat systolic BP waveforms from vagal activities using convolutional neural networks (CNNs). Three CNN models with different inputs and network structures were separately trained and tested for each rat including (1) Time–Frequency Sub-band CNN (TFS-CNN) with decomposed LVN sub-bands post wavelet transform as input and 1D convolutional layers, (2) Spectrogram-Spectral CNN (SS-CNN) with the 2D time–frequency spectrogram post Short Time Fourier Transform as input and 2D convolutional layers, and (3) Dilated Causal CNN (DC-CNN) with the same input as that in TFS-CNN and seven 1D dilated convolutional layers. It was found that the SS-CNN outperformed the other two models considering the reconstruction performance of BPs. The R² of SS-CNN (0.78 ± 0.08, mean<span><math><mo>±</mo></math></span>sd) exceeded those of TFS-CNN (0.70 ± 0.13) (P &lt; 0.05) and DC-CNN (0.53 ± 0.26) (P &lt; 0.05). The SS-CNN model also presented the least training (4.83 ± 1.98 mmHg) and testing loss (9.91 ± 3.89 mmHg). These findings would shed some light on clinical application of the closed-loop VNS in BP regulation.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108140"},"PeriodicalIF":4.9,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144281021","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A graph-based spine segmentation method: Combining target detection with unsupervised segmentation","authors":"Cong Zhang ,&nbsp;Kunjin He ,&nbsp;Wei Xu ,&nbsp;Xiaoqing Gu ,&nbsp;Zhengming Chen","doi":"10.1016/j.bspc.2025.108137","DOIUrl":"10.1016/j.bspc.2025.108137","url":null,"abstract":"<div><div>Spine image segmentation is important for the diagnosis and treatment of spinal diseases. However automatic segment vertebrae and intervertebral discs from spine images without segmentation labels is a challenging. In this paper, we propose an end-to-end spine image segmentation framework to achieve automated spine image segmentation. The framework consists of an initialization stage, a coarse segmentation stage and a fine segmentation stage. The initialization stage is a trained regions of interest detector. The coarse segmentation stage is a deep autoencoder clustering network. In this stage, the reconstruction loss and clustering loss of is used to achieve unsupervised coarse segmentation. In addition, a fixed number of channels strategy is also employed to greatly reduce the number of model parameters while ensuring the segmentation performance. In the fine segmentation stage, the image segmentation is reinterpreted from the perspective of the graph structure. The edge pixels from the coarse segmentation are used to construct graphs. The features of the nodes and the features of the edges between nodes are fully considered by the graph attention mechanism to achieve unsupervised fine segmentation. Experiments on two spine image segmentation datasets and one brain tumor image segmentation dataset show that our method has superior performance and generalization ability.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108137"},"PeriodicalIF":4.9,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144281020","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Detection, localization, segmentation, and classification in colorectal cancer screening using deep learning: A systematic review","authors":"Md. Rakibul Islam ,&nbsp;Md. Faysal Ahamed ,&nbsp;Md.Rabiul Islam ,&nbsp;Md. Nahiduzzaman ,&nbsp;Mominul Ahsan","doi":"10.1016/j.bspc.2025.108202","DOIUrl":"10.1016/j.bspc.2025.108202","url":null,"abstract":"<div><div>Colorectal polyps can develop into colorectal cancer (CRC), one of the leading causes of cancer-related deaths. These polyps must be found and treated as soon as possible to avoid developing CRC. Artificial intelligence (AI), especially deep learning (DL), has markedly improved the early diagnosis and treatment of CRC by increasing the accuracy and efficiency of polyp identification and classification. However, the rapid growth of research in this area has created a fragmented environment, highlighting the need for a comprehensive synthesis of findings to guide future advancements. This review aims to address this gap by systematically analyzing the state-of-the-art DL methodologies applied to colorectal polyp analysis. Adhering to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines, we conducted a structured search of major databases, identifying and analyzing 160 papers published between 2017 and 2024. The review focuses on the architectures of DL models, publicly available datasets, and performance metrics. The authors have analysed a wide range of DL architectures, including Convolutional Neural Networks (CNNs), YOLO-based object detectors, transformer models, recurrent neural networks (RNNs), autoencoders, and hybrid systems. Notably, YOLOv4 and CA-ResNet50 displayed state-of-the-art detection performance with accuracy rates exceeding 99 % on benchmark datasets such as LC25000 and Kvasir-SEG. For segmentation tasks, transformer-enhanced models like ViT and SwinE-Net earned excellent Dice scores (&gt;0.92), beating standard UNet variations. However, the research also reveals persisting problems, such as dataset variability, low generalizability across diverse populations, and the absence of standardized benchmarking techniques. In addition, we critically investigate the significance of data augmentation strategies in reducing dataset limits and overfitting, and also analyze how dataset characteristics affect model performance. Finally, structured future recommendations have been provided across three critical dimensions: (1) designing comprehensive and diverse datasets to improve generalizability, (2) establishing standardized and task-specific data augmentation pipelines, and (3) developing hybrid and modular architectures optimized for specific diagnostic subtasks. This thorough analysis aims to clarify the present status of research and highlight possibilities for enhancement in this vital healthcare sector.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108202"},"PeriodicalIF":4.9,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144280378","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-scale feature adaptive aggregation Transformer for super-resolution of lung computed tomography images","authors":"Yanmei Li,&nbsp;Qibin Yang,&nbsp;Fen Zhao,&nbsp;Jingshi Deng,&nbsp;Quanhao Ren,&nbsp;Yulong Pan","doi":"10.1016/j.bspc.2025.108126","DOIUrl":"10.1016/j.bspc.2025.108126","url":null,"abstract":"<div><div>High-resolution computed tomography (CT) images help doctors diagnose lung diseases by providing detailed information about underlying pathology. However, most current super-resolution methods still face the following problems: (1) Insufficient performance in restoring fine structure and high-frequency details of local edges, resulting in blurring of the reconstructed CT images. (2) These models are usually complex in structure and have a large number of parameters, which is both inefficient and requires additional computational resources. To address these issues, we propose an efficient Transformer model for super-resolution of lung CT images, named MFAT. Specifically, we propose a multi-scale feature adaptive aggregation strategy (MSAS) that splits features into multiple scales and uses independent computation at each scale to learn the corresponding feature representations while extracting image features in different receptive fields to enhance the fusion between the multi-level information. Additionally, we propose hybrid channel local window attention, which combines local context information and channel mixing to improve image texture expression and enhance detail clarity in reconstructed CT images. Finally, we design parameter-free attention mechanisms that utilize edge operators and multi-scale weighting to enhance highly contributing information and suppress redundant information, while also balancing the number of parameters. Extensive experiments on the COVID-CT dataset demonstrate that MFAT achieves a PSNR of 35.61 dB and 33.34 dB, and an SSIM of 0.9139 and 0.8706 at scale factors of ×3 and ×4, respectively, outperforming state-of-the-art methods. These results show that our method excels at reconstructing high-resolution lung CT images and recovering sharper image details.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108126"},"PeriodicalIF":4.9,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144280380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Annotation-free method for breast tumour segmentation in dynamic contrast-enhanced MRI","authors":"Jiahui He ,&nbsp;Junjie Zhang ,&nbsp;Xu Huang ,&nbsp;Yue Liu ,&nbsp;Jiayi Liao ,&nbsp;Yanfen Cui ,&nbsp;Wenbin Liu ,&nbsp;Changhong Liang ,&nbsp;Zaiyi Liu ,&nbsp;Lei Wu ,&nbsp;Gang Fang","doi":"10.1016/j.bspc.2025.108122","DOIUrl":"10.1016/j.bspc.2025.108122","url":null,"abstract":"<div><div>Breast cancer remains the leading cause of cancer-related mortality among women worldwide. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is a crucial tool for diagnosing breast tumours, as it provides high-resolution images of tissue structures. However, current methods for analysing these scans face two major limitations: they either require labour-intensive manual annotations by radiologists or rely on automated techniques with limited accuracy. To address these challenges, this study proposes an annotation-free method called the ACM (Algorithmic segmentation, Constraint filtering, Model training) that both eliminates the need for manual annotations and increases the segmentation accuracy. Unlike previous methods, which rely solely on deep learning or traditional algorithms, the ACM incorporates strict constraints to filter the pseudo-labels generated by traditional algorithms and integrates these pseudo-labels with unlabelled data for semi-supervised model training. Additionally, novel strategies, including the multi-class strategy and special augmentations, are introduced to mitigate common challenges in AI-based medical image analysis. Experimental validation on a large multicentre dataset comprising 1209 cases demonstrates that our method achieves a Dice similarity coefficient (DSC) of 83.06% in terms of tumour segmentation, approaching the performance of supervised methods. Furthermore, on an external test set, our method attains an Intersection over Union (IoU) of 74.45%, surpassing the best existing unsupervised methods by 23.4%. These results highlight the robustness and effectiveness of the ACM, which has the potential to significantly reduce the workload of radiologists, improve diagnostic consistency, and facilitate earlier breast cancer detection and personalized treatment planning. The source code is publicly available at <span><span>https://github.com/Ho-Garfield/ACM-pipeline</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108122"},"PeriodicalIF":4.9,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144280377","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CS-SwinGAN: A swin-transformer-based generative adversarial network with compressed sensing pre-enhancement for multi-coil MRI reconstruction","authors":"Haikang Zhang ,&nbsp;Zongqi Li ,&nbsp;Qingming Huang ,&nbsp;Luying Huang ,&nbsp;Yicheng Huang ,&nbsp;Wentao Wang ,&nbsp;Bing Shen","doi":"10.1016/j.bspc.2025.108120","DOIUrl":"10.1016/j.bspc.2025.108120","url":null,"abstract":"<div><div>Magnetic resonance imaging (MRI) reconstruction from undersampled k-space data is a crucial area of research due to its potential to reduce scan times. Current deep learning approaches for MRI reconstruction often combine frequency-domain and image-domain losses, optimizing their sum. However, this approach can lead to blurry results, as it averages two fundamentally different types of losses. To address this issue, we propose CS-SwinGAN for multi-coil MRI reconstruction, a swin-transformer-based generative adversarial network with a Compressed Sensing Block for pre-enhancement. The newly introduced Compressed Sensing Block not only facilitates the separation of frequency-domain and image-domain losses but also serves as a pre-enhancement stage that promotes sparsity and suppresses aliasing, thereby enhancing reconstruction quality. We evaluate CS-SwinGAN in both standard MRI reconstruction tasks and under varying noise levels in k-space to assess its performance across diverse conditions. Numerical experiments demonstrate that our framework outperforms state-of-the-art methods in both conventional reconstruction and noise suppression scenarios. The source code is available at <span><span>https://github.com/notmayday/CS-SwinGAN_MC_Rec</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108120"},"PeriodicalIF":4.9,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144272289","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dual-path network with dual-domain fusion and cross-attention for MRI reconstruction","authors":"Wenqi Chen ,&nbsp;Zhirong Gao ,&nbsp;Yuan He ,&nbsp;Jingxuan Wanyan ,&nbsp;Chengyi Xiong","doi":"10.1016/j.bspc.2025.108181","DOIUrl":"10.1016/j.bspc.2025.108181","url":null,"abstract":"<div><div>Accelerated magnetic resonance imaging (MRI) involves mapping the under-sampled k-space representation to reconstruct a high-quality image, and it remains a central challenge in MRI reconstruction. In recent years, deep learning has significantly improved the reconstruction performance in MRI. However, there is still a concern in the field regarding how to enhance the global feature learning ability of deep networks to further improve the quality of reconstructed images. To address this issue, this paper proposes a novel MRI reconstruction model called DDFCA-Net, which is based on a dual-path network with dual-domain fusion and cross-attention. The proposed DDFCA-Net model consists of two parallel and interactive paths. One path utilizes a convolutional neural network (CNN) to extract deep features from the k-space domain, while the other path employs a vision transformer (ViT) to extract deep features from the image domain. The image domain features and k-space features are mutually enhanced through cross-fusion. In the proposed model, the ViT network adopts a cross-attention learning strategy. The query matrix is derived from the image domain feature, while both the key matrix and value matrix are obtained from the fusion of the image domain feature and the k-space feature. This cross-attention mechanism promotes effective interaction between the two domains, further enhancing the feature extraction ability. Extensive experimental results on the CC359-Brain and FastMRI single-coil knee datasets, on different sampling rates and strategies, validate the effectiveness of the proposed method. The DDFCA-Net outperforms state-of-the-art methods, demonstrating superior reconstruction performance in MRI.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108181"},"PeriodicalIF":4.9,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144280379","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An end-to-end deep learning framework for the automated diagnosis of OPLL in CT images","authors":"Xiaolei Li,&nbsp;Duanwei Ma,&nbsp;Hao Zhang,&nbsp;Xiao Jia,&nbsp;Chuanpeng Li,&nbsp;Ran Song,&nbsp;Wei Zhang","doi":"10.1016/j.bspc.2025.108150","DOIUrl":"10.1016/j.bspc.2025.108150","url":null,"abstract":"<div><div>Accurate assessment of K-line status is vital for surgical planning and prognosis in patients with cervical Ossification of the Posterior Longitudinal Ligament (OPLL). However, variations in tissue appearance and morphological similarities between adjacent vertebrae in lateral cervical CT images complicate reliable edge identification for algorithms, posing challenges in recognizing easily confused vertebral landmarks. To address this issue, we propose an innovative approach that integrates K-line theory with deep learning techniques to evaluate K-line status efficiently and accurately in lateral cervical CT images of OPLL patients. Our method, the Dilated TransUNet, employs dilated convolution and Transformer modules to enhance the identification of easily confused vertebral landmarks, thereby improving detection accuracy. Additionally, we developed a discriminative algorithm utilizing dynamic thresholds for detailed pixel analysis around suspected ossification areas, effectively differentiating ossification from surrounding tissues. Experimental results demonstrate that our method achieves an average landmark detection accuracy of 98.49% and an image classification accuracy of 97.8%, both of which surpass existing methodologies. This framework reliably determines the position of ossification relative to the K-line, providing essential support for clinical surgical decision-making.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108150"},"PeriodicalIF":4.9,"publicationDate":"2025-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144262927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CFGMamba: Cross frame group Mamba for video-based depression recognition","authors":"Jingyi Liu ,&nbsp;Yuanyuan Shang ,&nbsp;Mengyuan Yang ,&nbsp;Zhuhong Shao ,&nbsp;Hui Ding ,&nbsp;Tie Liu","doi":"10.1016/j.bspc.2025.108113","DOIUrl":"10.1016/j.bspc.2025.108113","url":null,"abstract":"<div><div>Depression recognition is a significant research topic in the field of affective computing, which has important value for promoting clinical diagnosis and screening of depression. Video-based depression recognition methods utilize Convolutional Neural Networks (CNNs) or Transformers to capture relevant visual features and achieve promising performance. However, the limited receptive field of CNNs, the high computational resource consumption of Transformer long sequence modeling, and the high dimensionality of video data are key issues to be addressed. Considering these factors, this work introduces the State Space Model (SSM) for depression recognition and proposes a Cross Frame Group Mamba (CFGMamba) framework. CFGMamba alleviates the limitations of CNNs through global receptive fields and can effectively model long-range sequences with linear complexity. Technically, CFGMamba models cross-frame grouping of video data, dividing video frames into several distinct groups at time intervals and then performing bidirectional scanning for each group in the spatial–temporal dimension. This cross-frame grouping strategy efficiently captures richer emotional features while minimizing computational overhead. Meanwhile, CFGMamba incorporates a multi-stage downsampling approach, where multiple CFGMamba blocks are stacked at each stage to progressively capture multi-scale spatial–temporal emotional features from shallow to deep layers. Experimental results on the AVEC 2013 and AVEC 2014 datasets indicate that CFGMamba achieves competitive performance, with MAE/RMSE of 6.01/7.59 and 5.96/7.52, respectively. And the F1-score/AUC is 0.75/0.78 on the EmoReact dataset.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108113"},"PeriodicalIF":4.9,"publicationDate":"2025-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144272585","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}