Biomedical Signal Processing and Control最新文献_第5页

Multi-task learning for multi-scale breast cancer ultrasound image segmentation and classification based on visual perception 基于视觉感知的多任务学习多尺度乳腺癌超声图像分割与分类

IF 4.9 2区医学

Biomedical Signal Processing and Control Pub Date : 2025-06-18 DOI: 10.1016/j.bspc.2025.108212

Ao Su , Xiaolin Wang , Hao Xu , Jianfeng Zhang , Kang Chen , Dexing Kong , Guangfei Li , Xiaojun Chen , Jianming Wen , Zhong Lv

{"title":"Multi-task learning for multi-scale breast cancer ultrasound image segmentation and classification based on visual perception","authors":"Ao Su , Xiaolin Wang , Hao Xu , Jianfeng Zhang , Kang Chen , Dexing Kong , Guangfei Li , Xiaojun Chen , Jianming Wen , Zhong Lv","doi":"10.1016/j.bspc.2025.108212","DOIUrl":"10.1016/j.bspc.2025.108212","url":null,"abstract":"<div><div>Breast tumor segmentation and classification are essential components of breast ultrasound (BUS) computer-aided diagnosis (CAD) systems, which help improve the accuracy of breast cancer diagnoses. However, challenges arise due to the complexity of tumor features, the intensity similarities between lesions and surrounding tissues, as well as variations in tumor shape and location. While deep learning has been widely applied in CAD systems, many existing approaches overlook the relationship between segmentation and classification tasks, limiting their effectiveness. We propose PMSMT-Net, a Perception-based Multi-scale Ultrasound Image Segmentation and Classification Multi-task Learning Network, which enhances both segmentation and classification performance in BUS images. The segmentation network integrates a Visual Perception Module (VPM) to simulate human-like focus on regions of interest and, combined with Multi-scale Dilated Convolution (MSDC), accurately captures morphological, locational, and edge features. To further improve segmentation accuracy, we introduce the Variable Residual Convolutional Block Attention Module (VR-CBAM) and the Receptive Field Block-based Perceptually Separable Convolution Module (RFB-PSC), which enhance context feature fusion and reduce spatial information loss. The VPM output, along with the segmentation results, is then used as input to the classification network, where a transfer learning and ensemble learning approach classifies breast tumors. Compared to state-of-the-art methods, PMSMT-Net achieves an average improvement of 3.7% in Dice coefficient and 3.6% in classification accuracy on two public BUS datasets. These results demonstrate the proposed model can significantly advance BUS-based tumor analysis and is of great significance for improving diagnostic precision and patient outcomes.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108212"},"PeriodicalIF":4.9,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307055","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}

引用次数: 0

Rapid and accurate classification of retinal OCT diseases using reparameterized one-dimensional convolutional networks 用重参数化一维卷积网络快速准确地分类视网膜OCT疾病

IF 4.9 2区医学

Biomedical Signal Processing and Control Pub Date : 2025-06-18 DOI: 10.1016/j.bspc.2025.108167

Dexun Zhang , Mengjiao Zhang , Youming Sun , Wenjing Meng , Zhenzhen Li , Changmiao Wang , Huoling Luo , Zhengwei Zhang , Tianqiao Zhang

{"title":"Rapid and accurate classification of retinal OCT diseases using reparameterized one-dimensional convolutional networks","authors":"Dexun Zhang , Mengjiao Zhang , Youming Sun , Wenjing Meng , Zhenzhen Li , Changmiao Wang , Huoling Luo , Zhengwei Zhang , Tianqiao Zhang","doi":"10.1016/j.bspc.2025.108167","DOIUrl":"10.1016/j.bspc.2025.108167","url":null,"abstract":"<div><div>Efficient and precise classification of retinal optical coherence tomography (OCT) images is crucial for accurate diagnosis of eye diseases. However, existing classification networks often struggle with balancing inference speed and accuracy. To address this, we propose a novel retinal disease classification network that leverages prior knowledge of OCT images. By employing structural reparameterization and transforming the convolutional kernel shape to 1D, our network enhances its ability to focus on the inherent layering information of OCT images. Experimental results demonstrate that our approach significantly improves inference speed while maintaining high classification accuracy, compared to conventional and state-of-the-art networks. This advancement addresses real-time diagnostic needs in clinical settings. Our source code is available at: <span><span>https://github.com/xunlizhinian1124/1D-OCT-Classification</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108167"},"PeriodicalIF":4.9,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144306978","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}

引用次数: 0

DEQ-KAN: Deep equilibrium Kolmogorov–Arnold networks for robust classification 用于鲁棒分类的深度平衡Kolmogorov-Arnold网络

IF 4.9 2区医学

Biomedical Signal Processing and Control Pub Date : 2025-06-18 DOI: 10.1016/j.bspc.2025.108087

Jaber Qezelbash-Chamak

引用次数: 0

DR-CapsNet: deep residual capsule network with dynamic routing for automated identification of hepatocellular carcinoma and cirrhosis in CT images DR-CapsNet：基于动态路由的深度残留胶囊网络，用于CT图像中肝细胞癌和肝硬化的自动识别

IF 4.9 2区医学

Biomedical Signal Processing and Control Pub Date : 2025-06-18 DOI: 10.1016/j.bspc.2025.108201

Biao Qu , Wangfeng He , Xiaopeng Yao , Dongjing Shan , Jian Shu

{"title":"DR-CapsNet: deep residual capsule network with dynamic routing for automated identification of hepatocellular carcinoma and cirrhosis in CT images","authors":"Biao Qu , Wangfeng He , Xiaopeng Yao , Dongjing Shan , Jian Shu","doi":"10.1016/j.bspc.2025.108201","DOIUrl":"10.1016/j.bspc.2025.108201","url":null,"abstract":"<div><div>Hepatocellular carcinoma (HCC) diagnosis in CT images is challenging due to the overlapping imaging features with cirrhosis, particularly in early-stage small nodules, where rapid differentiation requires both morphological sensitivity and computational efficiency. To address this, we propose DR-CapsNet, a deep residual capsule network that combines lightweight residual blocks with dynamic routing mechanisms. The residual module alleviates gradient degradation through skip connections while enhancing the extraction of high-level image features, whereas the capsule framework leverages vector neurons and dynamic routing to model hierarchical part-whole relationships between cirrhotic nodules and HCC lesions. The dynamic routing mechanism iteratively refines coupling coefficients to establish affine-invariant spatial correlations across multi-scale capsule layers, allowing for precise differentiation of subtle morphological variations. Experimental results indicate that DR-CapsNet outperforms existing state-of-the-art methods in both accuracy and inference speed. Moreover, it exhibits exceptional robustness, even under limited training conditions (400 samples) and class imbalance (1:4 ratio) challenges. Overall, DR-CapsNet presents an accurate, efficient, and robust solution for the diagnosis of hepatocellular carcinoma (HCC), particularly in clinical settings with constrained resources.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108201"},"PeriodicalIF":4.9,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307054","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}

引用次数: 0

Human action recognition based on MnasNet optimized by improved version of Football Team training algorithm 改进版足球队训练算法优化的基于MnasNet的人体动作识别

IF 4.9 2区医学

Biomedical Signal Processing and Control Pub Date : 2025-06-18 DOI: 10.1016/j.bspc.2025.108207

Shiwen Lan, Yuan Xue, Huiping Liu, Xinfeng Yang

引用次数: 0

Nnet: N-type neural network for accurate segmentation of brain structures in MRI Nnet：用于MRI脑结构精确分割的n型神经网络

IF 4.9 2区医学

Biomedical Signal Processing and Control Pub Date : 2025-06-18 DOI: 10.1016/j.bspc.2025.108152

Xiufeng Zhang, Lingzhuo Tian, Yunfei Jiang, Shichen Zhang

{"title":"Nnet: N-type neural network for accurate segmentation of brain structures in MRI","authors":"Xiufeng Zhang, Lingzhuo Tian, Yunfei Jiang, Shichen Zhang","doi":"10.1016/j.bspc.2025.108152","DOIUrl":"10.1016/j.bspc.2025.108152","url":null,"abstract":"<div><div>Accurate segmentation of brain structures in magnetic resonance imaging plays a key role in the diagnosis and treatment of idiopathic Normal Pressure Hydrocephalus (iNPH). However, existing spatial domain-based methods are limited by the complexity of brain anatomical structures and the low contrast of medical images, making it difficult to achieve high-precision tissue boundary segmentation. To address this challenge, this paper re-examines the problem of brain structure segmentation from a new perspective in the frequency domain and proposes an N-type neural network architecture (Nnet). Nnet efficiently extracts low-frequency content information and high-frequency edge texture information in images through the coordinated operation of three parallel branches, thereby achieving accurate positioning of the target boundary. In addition, Nnet integrates the frequency domain online enhancement (FOE) module and the feature communication (FCM) module to further optimize the segmentation performance. The FOE module regulates the competition and cooperation between channels through the gating mechanism, effectively constructs the relationship between frequency domain features and reduces local information distortion. The FCM module uses the “growth” and “communication” mechanisms to fundamentally improve feature consistency by placing high-level and low-level features at the same level for communication, eliminating the semantic gap problem between branches. The results of this paper on two public brain MRI T1 sequence datasets (MALC dataset and IBSR dataset) show that the Dice coefficient of Nnet is improved by 1.29% and 1.47% on average, respectively, which is significantly better than the current advanced methods.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108152"},"PeriodicalIF":4.9,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144312670","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}

引用次数: 0

Automatic landmark detection and angle measurement in radiographs based on deep learning in application to coronal plane alignment of the knee classification 基于深度学习的x线片自动地标检测与角度测量在膝关节冠状面对齐分类中的应用

IF 4.9 2区医学

Biomedical Signal Processing and Control Pub Date : 2025-06-18 DOI: 10.1016/j.bspc.2025.108155

Xinru Zhong , Zhiyong Zhang , Hang Fang

{"title":"Automatic landmark detection and angle measurement in radiographs based on deep learning in application to coronal plane alignment of the knee classification","authors":"Xinru Zhong , Zhiyong Zhang , Hang Fang","doi":"10.1016/j.bspc.2025.108155","DOIUrl":"10.1016/j.bspc.2025.108155","url":null,"abstract":"<div><div>The Coronal Plane Alignment of the Knee (CPAK) based on radiographs is essential both for both preoperative planning and postoperative evaluation of knee arthroplasty. However, challenges in accurately detecting knee joint landmarks lead to imprecise knee-related angles measurement results, which in turn result in unreliable CPAK classification. In this paper, we propose an automated method for detecting landmarks, calculating knee-related angles, and performing CPAK classification using bilateral lower limb radiographs. Specifically, we construct a dataset and retrain a YOLO-based network to identify candidate regions for landmark detection. Landmark detection in radiographs is challenging due to the complexity of image details. To enhance the network’s ability to leverage important features, we introduce a Dual-path Fusion Attention Module, which uses Spatial Transformer Networks to focus on the skeletal region, and employs an Efficient Channel Attention Module to enhance edge features. A Coordinate Correction Module is proposed to facilitate multi-scale feature interaction, enabling accurate landmark localization. With precise landmark detection, our model achieves reliable angle measurement and CPAK classification. Extensive experiments demonstrate the superior performance of our network. The mean absolute errors for hip-knee-ankle angle, mechanical lateral distal femoral angle, mechanical medial proximal tibia angle and joint line convergence angle were 0.18°, 0.33°, 0.75° and 0.80° respectively. The intraclass correlation coefficients for all four angles were above 0.9.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108155"},"PeriodicalIF":4.9,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307053","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}

引用次数: 0

Cognitive load detection using adaptive/fixed-frequency empirical wavelet transform and multi-domain feature optimization 基于自适应/定频经验小波变换和多域特征优化的认知负荷检测

IF 4.9 2区医学

Biomedical Signal Processing and Control Pub Date : 2025-06-18 DOI: 10.1016/j.bspc.2025.108124

Jammisetty Yedukondalu , Lakhan Dev Sharma , Abhijit Bhattacharyya

{"title":"Cognitive load detection using adaptive/fixed-frequency empirical wavelet transform and multi-domain feature optimization","authors":"Jammisetty Yedukondalu , Lakhan Dev Sharma , Abhijit Bhattacharyya","doi":"10.1016/j.bspc.2025.108124","DOIUrl":"10.1016/j.bspc.2025.108124","url":null,"abstract":"<div><div>Neuronal activity is stimulated by cognitive load, which is crucial for comprehending the brain’s response to mental strain or stress-inducing stimuli. This study aims to look into the feasibility of extracting and classifying features from adaptive and traditional (rhythms) subbands of electroencephalogram (EEG) signals to assess cognitive load. Each EEG channel data (4-second duration) was decomposed into five subband signals (SBSs) using the empirical wavelet transform (EWT)-based multi-resolution analysis with adaptive and fixed spectral boundary frequencies. In fixed-frequency EWT (FF-EWT), the filter bank was designed with specific boundary frequencies for EEG rhythm extraction. In contrast to FF-EWT, the spectral boundaries of adaptive frequency EWT (AF-EWT) were adaptively found using the scale-space method. In the next step, multi-domain features (time-domain, frequency-domain, and non-linear features) were extracted from each EEG rhythm or SBS. The feature space dimension was reduced using binary atom search optimization (BASO) and binary equilibrium optimization (BEO) algorithms for improved classification performance. We have carried out a comprehensive study that includes feature-wise, rhythm/SBS-wise, and overall feature classification using seven machine learning techniques and their variants. Our proposed method combining FF-EWT-based multi-domain features with BASO and ensemble learning (EL) classifiers achieved the highest classification accuracy of 97.9%, 94.7%, and 99.1% in detecting cognitive loads using the mental arithmetic task (MAT), simultaneous workload (STEW), and mental load recognition (MLR) EEG datasets, respectively. Among rhythms and SBSs, the gamma rhythm appeared to play a significant role in analyzing a variety of cognitive tasks and achieved the highest classification accuracy. The proposed method outperformed the existing state-of-the-art techniques in the literature for cognitive load detection.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108124"},"PeriodicalIF":4.9,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144306977","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}

引用次数: 0

Development of a non-attached multi-person emotion recognition system based on sitting body motion signals 基于坐姿身体运动信号的非附着式多人情绪识别系统的研制

IF 4.9 2区医学

Biomedical Signal Processing and Control Pub Date : 2025-06-17 DOI: 10.1016/j.bspc.2025.108165

Hairui Fang , Yanpeng Ji , ShengLin Yuan , Genmin Qiu , Haoze Li , Zixuan Zhang , Lina Zhou

{"title":"Development of a non-attached multi-person emotion recognition system based on sitting body motion signals","authors":"Hairui Fang , Yanpeng Ji , ShengLin Yuan , Genmin Qiu , Haoze Li , Zixuan Zhang , Lina Zhou","doi":"10.1016/j.bspc.2025.108165","DOIUrl":"10.1016/j.bspc.2025.108165","url":null,"abstract":"<div><div>Automatic emotion recognition is a popular application field under the rapid development of sensing and information technology. Traditional automatic emotion recognition mainly uses attached sensors for target signal recognition, which is difficult to widely apply in practice because of expensive equipment, complex sensing systems, susceptibility to interference, and the potential to cause predictive emotional interference for users. In this work, a Non-attached Multi-person Emotion Recognition System (NMERS) is constructed, which recognizes the different emotions of multiple people through human sitting body motion signals. A flexible sensing cushion composed of four pressure sensors is used to collect real-time sitting motion signals from the user in a non-attached way. The Convolutional Neural Network-Long Short-Term Memory-Attention (CLATT) neural network model establishes the mapping relationship between the sitting body motion signals and four emotional categories, achieving an average accuracy of 98.05% for individual classification and 91.25% for cross-individual classification. Subsequently, by deploying CLATT on the upper computer and networking multiple sensing subsystems, real-time emotional state recognition of multi-person via a non-attached sensing system is successfully realized. NMERS builds a bridge between human body motion signals and emotional states, enhancing the practical application value of automatic emotion recognition systems.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108165"},"PeriodicalIF":4.9,"publicationDate":"2025-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144298001","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}

引用次数: 0

Effect of MAPT gene variations on the brain electrical activity: A multiplex network study MAPT基因变异对脑电活动的影响：一项多重网络研究

IF 4.9 2区医学

Biomedical Signal Processing and Control Pub Date : 2025-06-17 DOI: 10.1016/j.bspc.2025.108129

Aarón Maturana-Candelas , Roberto Hornero , Jesús Poza , Víctor Rodríguez-González , Víctor Gutiérrez-de Pablo , Nadia Pinto , Miguel A. Rebelo , Carlos Gómez

{"title":"Effect of MAPT gene variations on the brain electrical activity: A multiplex network study","authors":"Aarón Maturana-Candelas , Roberto Hornero , Jesús Poza , Víctor Rodríguez-González , Víctor Gutiérrez-de Pablo , Nadia Pinto , Miguel A. Rebelo , Carlos Gómez","doi":"10.1016/j.bspc.2025.108129","DOIUrl":"10.1016/j.bspc.2025.108129","url":null,"abstract":"<div><div>The aim of this study is to examine how variations in the microtubule-associated protein tau (<em>MAPT</em>) gene affect the brain functional network. For this purpose, resting-state electroencephalogram (EEG) data from 155 participants were acquired. This database included healthy controls and Alzheimer’s disease patients carrying seven <em>MAPT</em> alleles associated with risk or protective effects against neuropathologies or abnormal tau levels. To assess the impact of each genotype on brain function, a multiplex network analysis quantified the connectivity contribution of each brain region across multiple EEG frequency bands: (delta, theta, alpha, and beta). To this end, brain functional connectivity was first calculated for each brain region and frequency band using the phase lag index (PLI) parameter. The PLI adjacency matrices in each frequency band corresponded to the layers conforming the multiplex network. Subsequently, the participation coefficient (<em>P</em>) was computed in each brain region to reflect node degree diversification among frequency bands. Carriers of risk and protective alleles exhibited distinct values of <em>P</em>, especially in the left default mode network in healthy controls. In addition, carriers of the risk alleles generally presented higher network disruptions. Finally, significant differences in node degree values were observed across SNPs in the theta and beta frequency bands. These results suggest that different <em>MAPT</em> variants may lead to diverse tau species that influence brain function, particularly in brain regions involved in information flow management in preclinical states. These insights may help understanding network disturbances caused by molecular factors.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108129"},"PeriodicalIF":4.9,"publicationDate":"2025-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307052","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}

引用次数: 0