Biomedical Signal Processing and Control最新文献_第7页

GCN-multiDA: A multi-source personalized domain adaptation model based on a novel streamlined GCN for motor imagery classification GCN- multida：一种基于新型流线型GCN的多源个性化领域自适应运动图像分类模型

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2026-02-06 DOI: 10.1016/j.bspc.2026.109773

Zhenxi Zhao , Yingyu Cao , Hongbin Yu , Huixian Yu , Junfen Huang

Brain–computer interfaces (BCIs) play a pivotal role in facilitating human–machine interaction and elucidating brain mechanisms, with motor imagery (MI) being one of the most widely studied paradigms due to its substantial potential. However, inherent inter-subject variability in physiological structures often constrains the accuracy of MI decoding models. To address this challenge, we construct a streamlined graph convolutional network (GCN) and develop an MI decoding model, termed GCN-multiDA. Specifically, the model employs a GCN to capture spatial dependencies in EEG signals and incorporates a graph pruning strategy based on the task-frequency index (TF), region-of-interest index (ROI), and topological index (Topo) to streamline the network. This design preserves neurophysiological relevance while enhancing decoding accuracy and reducing model complexity. Furthermore, drawing inspiration from multi-source personalized domain adaptation, we introduce a domain bias assessment measurement (DBAM) to align cross-domain feature distributions and mitigate inter-domain discrepancies, along with a classifier alignment module to enforce prediction consistency across domains, thereby enabling robust MI classification. Comprehensive experiments conducted on four datasets, including BCI competition IV 2a and 2b, OpenBMI, and PhysioNet, demonstrate that GCN-multiDA consistently outperforms baseline models, improving mean accuracy by 2.66%, 2.53%, 1.32%, and 3.55%, respectively, and achieving the best performance in terms of Kappa and rRMSE metrics. Ablation and sensitivity analyses further confirm that the pruning algorithm contributes substantially to performance improvements across all datasets.

脑机接口（bci）在促进人机交互和阐明脑机制方面发挥着关键作用，其中运动意象（MI）因其巨大的潜力而成为研究最广泛的范式之一。然而，生理结构中固有的主体间可变性往往限制了MI解码模型的准确性。为了解决这一挑战，我们构建了一个流线型图卷积网络（GCN），并开发了一个MI解码模型，称为GCN- multida。具体而言，该模型采用GCN捕获EEG信号中的空间依赖关系，并结合基于任务频率指数（TF）、感兴趣区域指数（ROI）和拓扑指数（Topo）的图修剪策略来简化网络。该设计保留了神经生理学相关性，同时提高了解码精度并降低了模型复杂性。此外，从多源个性化领域自适应中获得灵感，我们引入了一个领域偏差评估测量（DBAM）来校准跨领域的特征分布并减轻领域间的差异，以及一个分类器校准模块来强制跨领域的预测一致性，从而实现鲁棒的MI分类。在BCI competition IV 2a和2b、OpenBMI和PhysioNet四个数据集上进行的综合实验表明，GCN-multiDA持续优于基线模型，平均准确率分别提高了2.66%、2.53%、1.32%和3.55%，并且在Kappa和rRMSE指标方面取得了最佳性能。消融和敏感性分析进一步证实，修剪算法对所有数据集的性能改进都有很大的贡献。

{"title":"GCN-multiDA: A multi-source personalized domain adaptation model based on a novel streamlined GCN for motor imagery classification","authors":"Zhenxi Zhao , Yingyu Cao , Hongbin Yu , Huixian Yu , Junfen Huang","doi":"10.1016/j.bspc.2026.109773","DOIUrl":"10.1016/j.bspc.2026.109773","url":null,"abstract":"<div><div>Brain–computer interfaces (BCIs) play a pivotal role in facilitating human–machine interaction and elucidating brain mechanisms, with motor imagery (MI) being one of the most widely studied paradigms due to its substantial potential. However, inherent inter-subject variability in physiological structures often constrains the accuracy of MI decoding models. To address this challenge, we construct a streamlined graph convolutional network (GCN) and develop an MI decoding model, termed GCN-multiDA. Specifically, the model employs a GCN to capture spatial dependencies in EEG signals and incorporates a graph pruning strategy based on the task-frequency index (TF), region-of-interest index (ROI), and topological index (Topo) to streamline the network. This design preserves neurophysiological relevance while enhancing decoding accuracy and reducing model complexity. Furthermore, drawing inspiration from multi-source personalized domain adaptation, we introduce a domain bias assessment measurement (DBAM) to align cross-domain feature distributions and mitigate inter-domain discrepancies, along with a classifier alignment module to enforce prediction consistency across domains, thereby enabling robust MI classification. Comprehensive experiments conducted on four datasets, including BCI competition IV 2a and 2b, OpenBMI, and PhysioNet, demonstrate that GCN-multiDA consistently outperforms baseline models, improving mean accuracy by 2.66%, 2.53%, 1.32%, and 3.55%, respectively, and achieving the best performance in terms of Kappa and rRMSE metrics. Ablation and sensitivity analyses further confirm that the pruning algorithm contributes substantially to performance improvements across all datasets.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"119 ","pages":"Article 109773"},"PeriodicalIF":4.9,"publicationDate":"2026-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146193099","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

RA2M-UNet: Efficient medical image segmentation via reparameterized convolution, dual-domain attention and 2D state–space modeling RA2M-UNet：通过重参数化卷积、双域关注和二维状态空间建模的高效医学图像分割

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2026-02-06 DOI: 10.1016/j.bspc.2026.109733

Chao Zhang , Lei Yang , Sai Zhang , Hongliang Duan , Jingjing Guo

Deep learning has made remarkable progress across various domains, particularly in medical image segmentation. However, a persistent challenge remains in balancing accuracy and computational efficiency, as current state-of-the-art models often sacrifice one aspect to enhance the other. Here, we propose RA2M-UNet, a novel network that addresses this trade-off through key innovations: (1) a feature fusion module that integrates multi-scale dilated convolutions with 2D selective scan module (2D-SSM); (2) an enhanced 2D-SSM for better spatial and semantic dependency capture; (3) parameter-efficient structural re-parameterization; (4) multi-output supervision for further refined segmentation. Comprehensive experiments demonstrate that our approach outperforms existing methods while maintaining parameter efficiency, effectively resolving the accuracy-efficiency dilemma in medical image segmentation.

深度学习在各个领域都取得了显著的进展，特别是在医学图像分割方面。然而，平衡准确性和计算效率仍然是一个持久的挑战，因为目前最先进的模型经常牺牲一个方面来增强另一个方面。在这里，我们提出了RA2M-UNet，这是一种通过关键创新解决这种权衡的新型网络：(1)将多尺度扩展卷积与2D选择性扫描模块（2D- ssm）集成在一起的特征融合模块；(2)增强2D-SSM，更好地捕获空间和语义依赖性；(3)参数高效结构再参数化；(4)多输出监督，进一步细化细分。综合实验表明，我们的方法在保持参数效率的前提下优于现有方法，有效解决了医学图像分割中精度-效率的难题。

引用次数: 0

Estimation of respiratory system parameters using an extended equation of motion during mandatory mechanical ventilation 在强制机械通气期间使用扩展运动方程估计呼吸系统参数

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2026-02-05 DOI: 10.1016/j.bspc.2026.109770

A. van Diepen , T.H.G.F. Bakkes , A.J.R. De Bie , R.A. Bouwman , P.H. Woerlee , M. Mischi , S. Turco

Mechanical ventilation is crucial for critically-ill patients, but requires continuous adjustments to prevent injuries and ensure optimal performance. Accurate estimation of respiratory parameters is essential for guiding disease-specific ventilator adjustments and optimizing respiratory support. The traditional equation of motion (EOM), widely used for this purpose, assumes linear compliance and resistance and neglects key physiological phenomena such as flow-dependent resistance and viscoelasticity, which can lead to biased estimates. In this study, we propose and validate an extended equation of motion (EEOM) that integrates nonlinear resistance and viscoelastic effects into a single framework. We fit the EEOM to simulated data, experimental data with a test lung, and clinical patient data (N=10). The accuracy of the EEOM to measure airway resistance, compliance, and viscoelasticity was improved compared to the traditional models (EOM and turbulent EOM, TEOM).

In simulations, EEOM method consistently reduced estimation errors across all scenarios. For example, in estimating compliance (

C_{rs}

), EEOM achieved an average percentage error range of 0.06-1.18% across all scenarios, compared to 2.5–14.5% with EOM and 10.5–18.6% with TEOM. Across all parameters and simulations, EEOM yielded lower total mean absolute errors (3.5%) than EOM (11.5%) and TEOM (11.9%). In clinical data, EEOM estimates for viscoelastic parameters (

C_{d}

,

R_{d}

) were statistically comparable to the interrupter technique (

p > 0.05

), additionally providing real-time estimates for nonlinear resistance terms.

These results demonstrate that extended mechanical parameters, including nonlinear resistance and viscoelasticity, can be identified noninvasively using standard ventilator signals, enabling more personalized, safer and adaptive ventilator settings.

机械通气对危重患者至关重要，但需要不断调整以防止受伤并确保最佳表现。准确估计呼吸参数对于指导特定疾病的呼吸机调整和优化呼吸支持至关重要。传统的运动方程（EOM）广泛用于此目的，它假设了线性顺应性和阻力，而忽略了关键的生理现象，如流动相关阻力和粘弹性，这可能导致有偏差的估计。在这项研究中，我们提出并验证了一个扩展的运动方程（EEOM），该方程将非线性阻力和粘弹性效应集成到一个单一的框架中。我们将EEOM与模拟数据、实验数据和临床患者数据（N=10）进行拟合。与传统模型（EOM和湍流EOM、TEOM）相比，EEOM测量气道阻力、顺应性和粘弹性的准确性得到了提高。在模拟中，EEOM方法一致地减少了所有场景的估计误差。例如，在评估符合性（Crs）时，EEOM在所有场景中实现的平均百分比误差范围为0.06-1.18%，而EOM为2.5-14.5%，TEOM为10.5-18.6%。在所有参数和模拟中，EEOM的总平均绝对误差（3.5%）低于EOM（11.5%）和TEOM（11.9%）。在临床数据中，EEOM对粘弹性参数（Cd, Rd）的估计在统计上与中断器技术相当（p>0.05），另外还提供了对非线性阻力项的实时估计。这些结果表明，扩展的机械参数，包括非线性阻力和粘弹性，可以使用标准呼吸机信号无创地识别，从而实现更个性化、更安全、更自适应的呼吸机设置。

{"title":"Estimation of respiratory system parameters using an extended equation of motion during mandatory mechanical ventilation","authors":"A. van Diepen , T.H.G.F. Bakkes , A.J.R. De Bie , R.A. Bouwman , P.H. Woerlee , M. Mischi , S. Turco","doi":"10.1016/j.bspc.2026.109770","DOIUrl":"10.1016/j.bspc.2026.109770","url":null,"abstract":"<div><div>Mechanical ventilation is crucial for critically-ill patients, but requires continuous adjustments to prevent injuries and ensure optimal performance. Accurate estimation of respiratory parameters is essential for guiding disease-specific ventilator adjustments and optimizing respiratory support. The traditional equation of motion (EOM), widely used for this purpose, assumes linear compliance and resistance and neglects key physiological phenomena such as flow-dependent resistance and viscoelasticity, which can lead to biased estimates. In this study, we propose and validate an extended equation of motion (EEOM) that integrates nonlinear resistance and viscoelastic effects into a single framework. We fit the EEOM to simulated data, experimental data with a test lung, and clinical patient data (N=10). The accuracy of the EEOM to measure airway resistance, compliance, and viscoelasticity was improved compared to the traditional models (EOM and turbulent EOM, TEOM).</div><div>In simulations, EEOM method consistently reduced estimation errors across all scenarios. For example, in estimating compliance (<span><math><msub><mrow><mi>C</mi></mrow><mrow><mi>rs</mi></mrow></msub></math></span>), EEOM achieved an average percentage error range of 0.06-1.18% across all scenarios, compared to 2.5–14.5% with EOM and 10.5–18.6% with TEOM. Across all parameters and simulations, EEOM yielded lower total mean absolute errors (3.5%) than EOM (11.5%) and TEOM (11.9%). In clinical data, EEOM estimates for viscoelastic parameters (<span><math><msub><mrow><mi>C</mi></mrow><mrow><mi>d</mi></mrow></msub></math></span>, <span><math><msub><mrow><mi>R</mi></mrow><mrow><mi>d</mi></mrow></msub></math></span>) were statistically comparable to the interrupter technique (<span><math><mrow><mi>p</mi><mo>></mo><mn>0</mn><mo>.</mo><mn>05</mn></mrow></math></span>), additionally providing real-time estimates for nonlinear resistance terms.</div><div>These results demonstrate that extended mechanical parameters, including nonlinear resistance and viscoelasticity, can be identified noninvasively using standard ventilator signals, enabling more personalized, safer and adaptive ventilator settings.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"118 ","pages":"Article 109770"},"PeriodicalIF":4.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174477","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

A novel unified complex network framework based on entropy moment for analyzing time series 一种新的基于熵矩的统一复杂网络框架用于时间序列分析

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2026-02-05 DOI: 10.1016/j.bspc.2026.109752

Ruiquan Chen , Jieren Xie , Yuqing Liu , Hanmin Chen , Junsheng Cheng , Shuo Tang , Yue Zhang , Xingxing Ke , Guanghua Xu , Bingwei He

Understanding electroencephalogram signals requires the exploration of nonlinear time series analysis techniques due to the intricate complexity of the human brain. Among these techniques, phase space entropy stands out, with bubble entropy recognized for its ability to mitigate the impact of selection parameters. However, various phase space entropies focus solely on the probability distribution of symbolized embedding vectors while disregarding the structural information and transition between symbols across different phase spaces. To address this limitation, we propose a novel definition of the entropy moment based on bubble entropy, termed the Bubble Transition Entropy Moment (BTEM). This enhancement allows for a better utilization of phase space information and introduces a new metric for assessing the regularity of time series. We conducted rigorous testing on a coupled Henon model to evaluate the efficacy of the proposed method. These tests highlighted its advantages in analyzing short-time series data and its resilience to fluctuations in parameters. In order to further validate the effectiveness of our method, we conducted experiments using two publicly epilepsy datasets. The results not only reaffirmed the superiority of the proposed unified framework over the traditional methods, but also demonstrated that it can achieve high decoding accuracy with shorter data lengths.

由于人类大脑的复杂性，理解脑电图信号需要探索非线性时间序列分析技术。在这些技术中，相空间熵脱颖而出，气泡熵因其减轻选择参数影响的能力而得到认可。然而，各种相空间熵只关注符号化嵌入向量的概率分布，而忽略了符号在不同相空间中的结构信息和转换。为了解决这一限制，我们提出了一个基于气泡熵的熵矩的新定义，称为气泡过渡熵矩（BTEM）。这种增强允许更好地利用相空间信息，并引入一种新的度量来评估时间序列的规律性。我们对耦合Henon模型进行了严格的测试，以评估所提出方法的有效性。这些测试突出了它在分析短时间序列数据和对参数波动的弹性方面的优势。为了进一步验证我们方法的有效性，我们使用两个公开的癫痫数据集进行了实验。结果不仅重申了所提出的统一框架相对于传统方法的优越性，而且证明了它可以在较短的数据长度下实现较高的解码精度。

{"title":"A novel unified complex network framework based on entropy moment for analyzing time series","authors":"Ruiquan Chen , Jieren Xie , Yuqing Liu , Hanmin Chen , Junsheng Cheng , Shuo Tang , Yue Zhang , Xingxing Ke , Guanghua Xu , Bingwei He","doi":"10.1016/j.bspc.2026.109752","DOIUrl":"10.1016/j.bspc.2026.109752","url":null,"abstract":"<div><div>Understanding electroencephalogram signals requires the exploration of nonlinear time series analysis techniques due to the intricate complexity of the human brain. Among these techniques, phase space entropy stands out, with bubble entropy recognized for its ability to mitigate the impact of selection parameters. However, various phase space entropies focus solely on the probability distribution of symbolized embedding vectors while disregarding the structural information and transition between symbols across different phase spaces. To address this limitation, we propose a novel definition of the entropy moment based on bubble entropy, termed the Bubble Transition Entropy Moment (BTEM). This enhancement allows for a better utilization of phase space information and introduces a new metric for assessing the regularity of time series. We conducted rigorous testing on a coupled Henon model to evaluate the efficacy of the proposed method. These tests highlighted its advantages in analyzing short-time series data and its resilience to fluctuations in parameters. In order to further validate the effectiveness of our method, we conducted experiments using two publicly epilepsy datasets. The results not only reaffirmed the superiority of the proposed unified framework over the traditional methods, but also demonstrated that it can achieve high decoding accuracy with shorter data lengths.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"118 ","pages":"Article 109752"},"PeriodicalIF":4.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174331","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

Cyclic deep representation-based domain adaptation for cross-subject motor imagery classification 基于循环深度表征的领域自适应跨主题运动意象分类

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2026-02-05 DOI: 10.1016/j.bspc.2026.109762

Min He , Xuan Cao , Tian-jian Luo

Deep representation learning has attracted great attention for brain-computer interfaces (BCIs) based neural rehabilitation engineering, especially for the motor imagery electroencephalogram (MI-EEG) signals. Recently researchers have explored numerous deep representation models to decode MI-EEG signals with various structures, however they suffered from the variability across recording subjects and the scarcity of samples. To solve these issues, domain adaptation models have been proposed to mitigate existed subjects’ samples to decode new subject’s sample by learning subject-invariant deep representations. However, existed models neglected temporal-varying and spatially-coupled characteristics of MI-EEG signals during domain adaptation, resulting performance deterioration for cross-subject classification. To improve decoding performance, we propose a novel domain adaptation model, referred to Cyclic Deep Representation-based Domain Adaptation (CDRDA), to simultaneously transfer deep representations from source domain to target domain, as well as target domain to source domain. Specifically, our CDRDA model learns a joint optimization that weighted dual adversarial losses, cyclic losses, and domain-specific losses to improve classification performance together. Empirical experiments on two benchmark MI-EEG datasets have revealed the feasibility and effectiveness of the CDRDA model with accuracy, Cohen’s kappa, and macro average F1-score. Results analyses and ablation studies have also verified the superiority of the CDRDA model for building online MI-BCIs.

深度表征学习在基于脑机接口（bci）的神经康复工程中备受关注，尤其是在运动图像脑电图（MI-EEG）信号方面。近年来，研究人员已经探索了许多深度表示模型来解码具有不同结构的MI-EEG信号，但它们受到记录对象的可变性和样本稀缺性的影响。为了解决这些问题，研究人员提出了领域适应模型，通过学习主体不变的深度表征来缓解已有主体的样本，从而解码新主体的样本。然而，现有模型在域自适应过程中忽略了脑电信号的时变和空间耦合特征，导致跨主题分类性能下降。为了提高解码性能，我们提出了一种新的基于循环深度表示的域自适应（CDRDA）模型，该模型可以同时将深度表示从源域传输到目标域，以及从目标域传输到源域。具体来说，我们的CDRDA模型学习了加权对偶对抗损失、循环损失和特定领域损失的联合优化，以共同提高分类性能。在两个基准MI-EEG数据集上的实证实验表明，CDRDA模型具有准确率、Cohen’s kappa和宏观平均f1得分的可行性和有效性。结果分析和消融研究也验证了CDRDA模型构建在线mi - bci的优越性。

{"title":"Cyclic deep representation-based domain adaptation for cross-subject motor imagery classification","authors":"Min He , Xuan Cao , Tian-jian Luo","doi":"10.1016/j.bspc.2026.109762","DOIUrl":"10.1016/j.bspc.2026.109762","url":null,"abstract":"<div><div>Deep representation learning has attracted great attention for brain-computer interfaces (BCIs) based neural rehabilitation engineering, especially for the motor imagery electroencephalogram (MI-EEG) signals. Recently researchers have explored numerous deep representation models to decode MI-EEG signals with various structures, however they suffered from the variability across recording subjects and the scarcity of samples. To solve these issues, domain adaptation models have been proposed to mitigate existed subjects’ samples to decode new subject’s sample by learning subject-invariant deep representations. However, existed models neglected temporal-varying and spatially-coupled characteristics of MI-EEG signals during domain adaptation, resulting performance deterioration for cross-subject classification. To improve decoding performance, we propose a novel domain adaptation model, referred to <strong>C</strong>yclic <strong>D</strong>eep <strong>R</strong>epresentation-based <strong>D</strong>omain <strong>A</strong>daptation (CDRDA), to simultaneously transfer deep representations from source domain to target domain, as well as target domain to source domain. Specifically, our CDRDA model learns a joint optimization that weighted dual adversarial losses, cyclic losses, and domain-specific losses to improve classification performance together. Empirical experiments on two benchmark MI-EEG datasets have revealed the feasibility and effectiveness of the CDRDA model with accuracy, Cohen’s kappa, and macro average F1-score. Results analyses and ablation studies have also verified the superiority of the CDRDA model for building online MI-BCIs.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"118 ","pages":"Article 109762"},"PeriodicalIF":4.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174332","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

Spatially Enhanced Pyramid Split attention for improved ECG-Based emotion recognition 空间增强金字塔分裂注意改进基于心电的情绪识别

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2026-02-05 DOI: 10.1016/j.bspc.2026.109729

Chenyang Pan , Hui Chen , Xuedian Zhang , Tong Su , Pei Ma

Accurate emotion recognition plays a crucial role in human–computer interaction, mental healthcare, and cognitive behavior analysis. Previous research consistently demonstrates ECG’s strong potential for emotion recognition, yet current ECG-based approaches still face critical challenges including sensitivity to signal artifacts, inter-subject variability, and limited discriminability in fine-grained emotion classification. To address these issues, we propose a deep learning framework that enhances feature representation through a Spatially Enhanced Pyramid Split-Attention (SEPSA) mechanism, which captures multi-scale spatial patterns in ECG signals to enable more robust emotion classification from raw inputs. The method further introduces an optimized beat-level preprocessing strategy to improve data quality by identifying and removing morphologically inconsistent heartbeats. Extensive experiments on two public datasets—WESAD and DREAMER—showed that our framework achieved competitive performance. It attained an average accuracy of 98.9% in four-class emotion classification on WESAD, and 94.5%/92.7% in binary classification of arousal and valence on DREAMER, where it also reached 89.8%/88.7% as the average accuracy in five-class classification. Ablation studies confirmed the contribution of each component to the overall performance. These results underscore the effectiveness of our approach within the studied datasets and suggest its feasibility as a foundation for future research. Subsequent work will focus on enhancing generalizability through validation on larger, more ecologically diverse datasets and exploring integration pathways for wearable affective computing systems.

准确的情绪识别在人机交互、心理健康和认知行为分析中起着至关重要的作用。先前的研究一致证明了ECG在情绪识别方面的强大潜力，但目前基于ECG的方法仍然面临着关键的挑战，包括对信号伪像的敏感性、主体间的可变性以及细粒度情绪分类的有限可辨析性。为了解决这些问题，我们提出了一个深度学习框架，该框架通过空间增强金字塔分散注意（SEPSA）机制增强特征表示，该机制捕获ECG信号中的多尺度空间模式，从而从原始输入中实现更稳健的情绪分类。该方法进一步引入了一种优化的心跳级预处理策略，通过识别和去除形态不一致的心跳来提高数据质量。在两个公共数据集（wesad和dreamer）上进行的大量实验表明，我们的框架取得了具有竞争力的性能。WESAD上情绪四级分类的平均正确率为98.9%，dream上唤醒效价二元分类的平均正确率为94.5%/92.7%，其中唤醒效价二元分类的平均正确率为89.8%/88.7%。消融研究证实了每个部件对整体性能的贡献。这些结果强调了我们的方法在研究数据集中的有效性，并表明其作为未来研究基础的可行性。后续工作将侧重于通过对更大、更生态多样化的数据集进行验证来增强通用性，并探索可穿戴情感计算系统的集成途径。

{"title":"Spatially Enhanced Pyramid Split attention for improved ECG-Based emotion recognition","authors":"Chenyang Pan , Hui Chen , Xuedian Zhang , Tong Su , Pei Ma","doi":"10.1016/j.bspc.2026.109729","DOIUrl":"10.1016/j.bspc.2026.109729","url":null,"abstract":"<div><div>Accurate emotion recognition plays a crucial role in human–computer interaction, mental healthcare, and cognitive behavior analysis. Previous research consistently demonstrates ECG’s strong potential for emotion recognition, yet current ECG-based approaches still face critical challenges including sensitivity to signal artifacts, inter-subject variability, and limited discriminability in fine-grained emotion classification. To address these issues, we propose a deep learning framework that enhances feature representation through a Spatially Enhanced Pyramid Split-Attention (SEPSA) mechanism, which captures multi-scale spatial patterns in ECG signals to enable more robust emotion classification from raw inputs. The method further introduces an optimized beat-level preprocessing strategy to improve data quality by identifying and removing morphologically inconsistent heartbeats. Extensive experiments on two public datasets—WESAD and DREAMER—showed that our framework achieved competitive performance. It attained an average accuracy of 98.9% in four-class emotion classification on WESAD, and 94.5%/92.7% in binary classification of arousal and valence on DREAMER, where it also reached 89.8%/88.7% as the average accuracy in five-class classification. Ablation studies confirmed the contribution of each component to the overall performance. These results underscore the effectiveness of our approach within the studied datasets and suggest its feasibility as a foundation for future research. Subsequent work will focus on enhancing generalizability through validation on larger, more ecologically diverse datasets and exploring integration pathways for wearable affective computing systems.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"118 ","pages":"Article 109729"},"PeriodicalIF":4.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174475","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

Explainable deep autoencoding of vibroarthrographic time–frequency distributions for robust knee disorder detection 可解释的深度自编码振动关节时间-频率分布鲁棒膝盖疾病检测

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2026-02-05 DOI: 10.1016/j.bspc.2026.109781

Saif Nalband , Maulik Gupta , Sachin Kansale , Tanmoy Hazra , Femi Robert , A. Amalin Prince

This paper presents AUTOENCODE-KNEE, a novel approach for automatic feature extraction from the time–frequency distribution of the knee joint of the human body using vibroarthrographic (VAG) signals. VAG signals contain valuable information, which is crucial for diagnosing various musculoskeletal disorders. However, manually extracting relevant features from VAG signals can be time-consuming and subjective. To address this challenge, we propose utilizing a convolutional neural network (CNN)-based autoencoder architecture for automatic feature extraction. The autoencoder is trained on a dataset comprising time–frequency representations of VAG signals, learning to encode and decode the input signals while preserving important features. By leveraging the inherent ability of CNNs to capture spatial dependencies, the autoencoder effectively learns to extract discriminative features from the complex time–frequency domain. Our experimental results demonstrate the efficacy of AUTOENCODE-KNEE in automatically extracting meaningful features from knee joint signals. We compare different machine learning models for classifying musculoskeletal disorders. Furthermore, we use explainable Artificial Intelligence (xAI) to capture more abstract and pathology-relevant features. In summary, AUTOENCODE-KNEE offers a promising solution for automatic feature extraction from knee joint signals, potentially revolutionizing how musculoskeletal disorders are diagnosed and treated.

本文提出了一种利用关节振动成像（VAG）信号从人体膝关节的时频分布中自动提取特征的新方法——AUTOENCODE-KNEE。VAG信号包含有价值的信息，对诊断各种肌肉骨骼疾病至关重要。但是，手动从VAG信号中提取相关特征费时且主观。为了解决这一挑战，我们提出利用基于卷积神经网络（CNN）的自编码器架构进行自动特征提取。自动编码器在包含VAG信号时频表示的数据集上进行训练，学习编码和解码输入信号，同时保留重要特征。通过利用cnn固有的捕获空间依赖关系的能力，自编码器有效地学习从复杂的时频域中提取判别特征。我们的实验结果证明了AUTOENCODE-KNEE在从膝关节信号中自动提取有意义特征方面的有效性。我们比较了用于分类肌肉骨骼疾病的不同机器学习模型。此外，我们使用可解释的人工智能（xAI）来捕获更抽象和病理相关的特征。总之，AUTOENCODE-KNEE为膝关节信号的自动特征提取提供了一个很有前途的解决方案，可能会彻底改变肌肉骨骼疾病的诊断和治疗方式。

{"title":"Explainable deep autoencoding of vibroarthrographic time–frequency distributions for robust knee disorder detection","authors":"Saif Nalband , Maulik Gupta , Sachin Kansale , Tanmoy Hazra , Femi Robert , A. Amalin Prince","doi":"10.1016/j.bspc.2026.109781","DOIUrl":"10.1016/j.bspc.2026.109781","url":null,"abstract":"<div><div>This paper presents AUTOENCODE-KNEE, a novel approach for automatic feature extraction from the time–frequency distribution of the knee joint of the human body using vibroarthrographic (VAG) signals. VAG signals contain valuable information, which is crucial for diagnosing various musculoskeletal disorders. However, manually extracting relevant features from VAG signals can be time-consuming and subjective. To address this challenge, we propose utilizing a convolutional neural network (CNN)-based autoencoder architecture for automatic feature extraction. The autoencoder is trained on a dataset comprising time–frequency representations of VAG signals, learning to encode and decode the input signals while preserving important features. By leveraging the inherent ability of CNNs to capture spatial dependencies, the autoencoder effectively learns to extract discriminative features from the complex time–frequency domain. Our experimental results demonstrate the efficacy of AUTOENCODE-KNEE in automatically extracting meaningful features from knee joint signals. We compare different machine learning models for classifying musculoskeletal disorders. Furthermore, we use explainable Artificial Intelligence (xAI) to capture more abstract and pathology-relevant features. In summary, AUTOENCODE-KNEE offers a promising solution for automatic feature extraction from knee joint signals, potentially revolutionizing how musculoskeletal disorders are diagnosed and treated.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"118 ","pages":"Article 109781"},"PeriodicalIF":4.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174482","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

Robust source-free few-shot brain tumor segmentation via style perturbation and heatmap-guided consistency 基于风格扰动和热图引导一致性的鲁棒无源少射脑肿瘤分割

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2026-02-05 DOI: 10.1016/j.bspc.2026.109790

Li Liu , Khairunnisa Hasikin , Kaijian Xia , Khin Wee Lai

Cross-domain few-shot segmentation (CD-FSS) has shown great potential for alleviating annotation scarcity and mitigating domain discrepancies in medical image analysis. However, existing CD-FSS approaches typically require access to source-domain data, which conflicts with clinical privacy regulations and data-isolation constraints. To address this issue, we propose SHCNet, a source-free few-shot brain tumor segmentation framework for MRI. SHCNet consists of three key components: (i) a feature-level style perturbation module (EnhanceStyle) that improves robustness to domain-specific appearance variations; (ii) a heatmap-guided structure alignment mechanism (HGSA) for enforcing spatial saliency consistency between support and query features; and (iii) a semantic consistency alignment (SCA) module that enhances intra-class compactness and inter-class separability via foreground–background distance constraints and triplet loss. Extensive experiments show that SHCNet significantly outperforms the state-of-the-art source-free method ABCDFSS across diverse datasets. On BraTS 2020, SHCNet (ResNet-50) achieves mean DSCs of 76.22% and 80.36% under 1-shot and 5-shot settings, yielding + 12.54 pp and + 16.23 pp improvements. On BraTS 2021, the gains reach + 14.87 pp and + 16.65 pp, while on BraTS Africa, SHCNet obtains + 8.27 pp and + 10.93 pp. Moreover, SHCNet delivers the lowest HD95 (down to 6.33 mm), demonstrating strong boundary awareness and cross-domain robustness. These results verify that SHCNet provides an effective solution for source-free few-shot tumor segmentation under clinically realistic constraints.

在医学图像分析中，Cross-domain few-shot segmentation （CD-FSS）在缓解注释稀缺性和缓解域差异方面显示出巨大的潜力。然而，现有的CD-FSS方法通常需要访问源域数据，这与临床隐私法规和数据隔离约束相冲突。为了解决这个问题，我们提出了SHCNet，一个无源的MRI小片段脑肿瘤分割框架。SHCNet由三个关键组件组成：(i)一个特征级风格扰动模块（enhancyle），它提高了对特定领域外观变化的鲁棒性；（ii）以热图为导向的结构对齐机制（HGSA），以加强支持和查询特征之间的空间显著性一致性；（iii）语义一致性对齐（SCA）模块，通过前景-背景距离约束和三元组丢失增强类内紧凑性和类间可分离性。大量的实验表明，在不同的数据集上，SHCNet显著优于最先进的无源方法ABCDFSS。在BraTS 2020上，SHCNet （ResNet-50）在1针和5针设置下的平均dsc分别为76.22%和80.36%，分别提高了12.54和16.23个pp。在BraTS 2021上，增益达到+ 14.87 pp和+ 16.65 pp，而在BraTS Africa上，SHCNet获得+ 8.27 pp和+ 10.93 pp。此外，SHCNet提供最低的HD95（降至6.33 mm），表现出强大的边界感知和跨域鲁棒性。这些结果验证了SHCNet在临床现实约束下为无源少次肿瘤分割提供了有效的解决方案。

{"title":"Robust source-free few-shot brain tumor segmentation via style perturbation and heatmap-guided consistency","authors":"Li Liu , Khairunnisa Hasikin , Kaijian Xia , Khin Wee Lai","doi":"10.1016/j.bspc.2026.109790","DOIUrl":"10.1016/j.bspc.2026.109790","url":null,"abstract":"<div><div>Cross-domain few-shot segmentation (CD-FSS) has shown great potential for alleviating annotation scarcity and mitigating domain discrepancies in medical image analysis. However, existing CD-FSS approaches typically require access to source-domain data, which conflicts with clinical privacy regulations and data-isolation constraints. To address this issue, we propose SHCNet, a source-free few-shot brain tumor segmentation framework for MRI. SHCNet consists of three key components: (i) a feature-level style perturbation module (EnhanceStyle) that improves robustness to domain-specific appearance variations; (ii) a heatmap-guided structure alignment mechanism (HGSA) for enforcing spatial saliency consistency between support and query features; and (iii) a semantic consistency alignment (SCA) module that enhances intra-class compactness and inter-class separability via foreground–background distance constraints and triplet loss. Extensive experiments show that SHCNet significantly outperforms the state-of-the-art source-free method ABCDFSS across diverse datasets. On BraTS 2020, SHCNet (ResNet-50) achieves mean DSCs of 76.22% and 80.36% under 1-shot and 5-shot settings, yielding + 12.54 pp and + 16.23 pp improvements. On BraTS 2021, the gains reach + 14.87 pp and + 16.65 pp, while on BraTS Africa, SHCNet obtains + 8.27 pp and + 10.93 pp. Moreover, SHCNet delivers the lowest HD95 (down to 6.33 mm), demonstrating strong boundary awareness and cross-domain robustness. These results verify that SHCNet provides an effective solution for source-free few-shot tumor segmentation under clinically realistic constraints.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"118 ","pages":"Article 109790"},"PeriodicalIF":4.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174336","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

AFET: Adaptive Frequency-Enhanced Transformer for X-ray image compression 用于x射线图像压缩的自适应频率增强变压器

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2026-02-05 DOI: 10.1016/j.bspc.2026.109727

Tiansong Li , Qingsong Yang , Guofen Wang , Shaoguo Cui , Hongkui Wang , Li Yu

With the rapid development of modern medical imaging equipment, X-ray image resolution continues to increase, leading to an exponential growth in data volume. To alleviate the storage and transmission burden of massive X-ray image data, efficient compression has become a key requirement for contemporary medical information systems. An adaptive frequency-enhancement transformer (AFET) is proposed for X-ray image compression by leveraging adaptive enhanced window attention (AEWA) and a novel deep feedforward network (DFFN) to build a multi-frequency domain interaction mechanism. Firstly, the AEWA module is designed to dynamically enhances key frequency components through adaptive weighting across windows, effectively eliminating redundancy and preserving subtle details. Secondly, the DFFN module is introduced to capture spatial correlations between different frequency components, improving structural feature extraction. Experiments on the ChestX-ray8 and CheXpert datasets demonstrate that AFET outperforms state-of-the-art learning-based compressors and traditional codecs (such as JPEG and BPG) in terms of PSNR and MS-SSIM metrics, achieving a BD-Rate reduction of 19.51% on ChestX-ray8. Furthermore, the clinical feasibility of our AFET was verified in a downstream fine-grained classification task on compressed images, achieving superior AUC scores in chest disease classification, confirming the effectiveness of our AFET in clinical applications. Code: https://github.com/TiansongLi/AFET.

随着现代医学成像设备的快速发展，x射线图像分辨率不断提高，导致数据量呈指数级增长。为了减轻海量x射线图像数据的存储和传输负担，高效压缩已成为当代医疗信息系统的关键要求。提出了一种用于x射线图像压缩的自适应频率增强变压器（AFET），该变压器利用自适应增强窗口注意（AEWA）和一种新型的深度前馈网络（DFFN）来构建多频域交互机制。首先，AEWA模块通过自适应跨窗口加权来动态增强关键频率分量，有效消除冗余并保留细微细节。其次，引入DFFN模块捕获不同频率分量之间的空间相关性，提高结构特征提取的效率；在ChestX-ray8和CheXpert数据集上的实验表明，在PSNR和MS-SSIM指标方面，AFET优于最先进的基于学习的压缩器和传统的编码器（如JPEG和BPG），在ChestX-ray8上实现了19.51%的BD-Rate降低。此外，在压缩图像的下游细粒度分类任务中验证了我们的AFET的临床可行性，在胸部疾病分类中获得了优异的AUC评分，证实了我们的AFET在临床应用中的有效性。代码:https://github.com/TiansongLi/AFET。

{"title":"AFET: Adaptive Frequency-Enhanced Transformer for X-ray image compression","authors":"Tiansong Li , Qingsong Yang , Guofen Wang , Shaoguo Cui , Hongkui Wang , Li Yu","doi":"10.1016/j.bspc.2026.109727","DOIUrl":"10.1016/j.bspc.2026.109727","url":null,"abstract":"<div><div>With the rapid development of modern medical imaging equipment, X-ray image resolution continues to increase, leading to an exponential growth in data volume. To alleviate the storage and transmission burden of massive X-ray image data, efficient compression has become a key requirement for contemporary medical information systems. An adaptive frequency-enhancement transformer (AFET) is proposed for X-ray image compression by leveraging adaptive enhanced window attention (AEWA) and a novel deep feedforward network (DFFN) to build a multi-frequency domain interaction mechanism. Firstly, the AEWA module is designed to dynamically enhances key frequency components through adaptive weighting across windows, effectively eliminating redundancy and preserving subtle details. Secondly, the DFFN module is introduced to capture spatial correlations between different frequency components, improving structural feature extraction. Experiments on the ChestX-ray8 and CheXpert datasets demonstrate that AFET outperforms state-of-the-art learning-based compressors and traditional codecs (such as JPEG and BPG) in terms of PSNR and MS-SSIM metrics, achieving a BD-Rate reduction of 19.51% on ChestX-ray8. Furthermore, the clinical feasibility of our AFET was verified in a downstream fine-grained classification task on compressed images, achieving superior AUC scores in chest disease classification, confirming the effectiveness of our AFET in clinical applications. Code: <span><span>https://github.com/TiansongLi/AFET</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"118 ","pages":"Article 109727"},"PeriodicalIF":4.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174338","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

From CNNs to diffusion models: a decade of advances in brain tumor and stroke lesion segmentation across BraTS, ATLAS, and ISLES benchmarks 从cnn到扩散模型：跨越BraTS、ATLAS和ISLES基准的脑肿瘤和中风病灶分割的十年进展

IF 4.9 2区医学 Q1 ENGINEERING, BIOMEDICAL

Biomedical Signal Processing and Control

Pub Date : 2026-02-05 DOI: 10.1016/j.bspc.2026.109780

Manzoor Mohammad, Burra Vijaya Babu

Accurate segmentation of brain tumors and ischemic stroke lesions from magnetic resonance imaging (MRI) is a fundamental step in diagnosis, prognosis, and treatment planning. Over the past decade, segmentation research has advanced rapidly with the development of deep learning architectures, large annotated datasets, and diverse evaluation protocols. This survey provides a unified overview of segmentation techniques spanning both brain tumors and ischemic stroke lesions, offering a comprehensive perspective on their evolution, strengths, and limitations. We examine key benchmark datasets, including BraTS (2015–2023), ISLES (2015–2022), and ATLAS v2.0, and summarize their imaging modalities, annotations, and clinical contexts. Major model families such as U-Net variants, CNN-Transformer hybrids, transformer-only models, ensemble frameworks, and emerging diffusion-based approaches are systematically analysed with respect to design principles and reported performance across subregions and lesion types. The survey further compiles leaderboard results, state-of-the-art comparisons, and over 70 influential studies, highlighting region-wise trends and performance variability. Visual analyses, including histograms and box plots, offer additional insight into how segmentation accuracy differs across datasets and methods. We also review persistent challenges such as data scarcity, domain shift, boundary ambiguity, and clinical integration barriers, along with research trends including foundation models, semi- and self-supervised learning, and multimodal fusion. By consolidating datasets, methodologies, evaluation metrics, and future directions, this survey serves as a reference point for researchers and clinicians, outlining both the progress made and the opportunities that remain in developing robust and clinically relevant segmentation systems.

从磁共振成像（MRI）中准确分割脑肿瘤和缺血性脑卒中病变是诊断、预后和治疗计划的基本步骤。在过去的十年中，随着深度学习架构、大型注释数据集和各种评估协议的发展，分割研究取得了快速进展。这项调查提供了一个统一的概述分割技术跨越脑肿瘤和缺血性中风病变，提供了一个全面的观点，他们的发展，优势和局限性。我们研究了关键的基准数据集，包括BraTS（2015-2023）、ISLES（2015-2022）和ATLAS v2.0，并总结了它们的成像方式、注释和临床背景。主要的模型家族，如U-Net变体、CNN-Transformer混合模型、仅变压器模型、集成框架和新兴的基于扩散的方法，系统地分析了设计原则和跨子区域和病变类型的报告性能。该调查进一步汇总了排行榜结果、最先进的比较和70多项有影响力的研究，突出了地区趋势和绩效差异。可视化分析，包括直方图和箱形图，提供了更多的洞察如何分割精度不同的数据集和方法。我们还回顾了持续的挑战，如数据稀缺，领域转移，边界模糊，临床整合障碍，以及研究趋势，包括基础模型，半和自我监督学习，以及多模态融合。通过整合数据集、方法、评估指标和未来方向，该调查为研究人员和临床医生提供了参考点，概述了在开发稳健的临床相关分割系统方面取得的进展和仍然存在的机会。

{"title":"From CNNs to diffusion models: a decade of advances in brain tumor and stroke lesion segmentation across BraTS, ATLAS, and ISLES benchmarks","authors":"Manzoor Mohammad, Burra Vijaya Babu","doi":"10.1016/j.bspc.2026.109780","DOIUrl":"10.1016/j.bspc.2026.109780","url":null,"abstract":"<div><div>Accurate segmentation of brain tumors and ischemic stroke lesions from magnetic resonance imaging (MRI) is a fundamental step in diagnosis, prognosis, and treatment planning. Over the past decade, segmentation research has advanced rapidly with the development of deep learning architectures, large annotated datasets, and diverse evaluation protocols. This survey provides a unified overview of segmentation techniques spanning both brain tumors and ischemic stroke lesions, offering a comprehensive perspective on their evolution, strengths, and limitations. We examine key benchmark datasets, including BraTS (2015–2023), ISLES (2015–2022), and ATLAS v2.0, and summarize their imaging modalities, annotations, and clinical contexts. Major model families such as U-Net variants, CNN-Transformer hybrids, transformer-only models, ensemble frameworks, and emerging diffusion-based approaches are systematically analysed with respect to design principles and reported performance across subregions and lesion types. The survey further compiles leaderboard results, state-of-the-art comparisons, and over 70 influential studies, highlighting region-wise trends and performance variability. Visual analyses, including histograms and box plots, offer additional insight into how segmentation accuracy differs across datasets and methods. We also review persistent challenges such as data scarcity, domain shift, boundary ambiguity, and clinical integration barriers, along with research trends including foundation models, semi- and self-supervised learning, and multimodal fusion. By consolidating datasets, methodologies, evaluation metrics, and future directions, this survey serves as a reference point for researchers and clinicians, outlining both the progress made and the opportunities that remain in developing robust and clinically relevant segmentation systems.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"118 ","pages":"Article 109780"},"PeriodicalIF":4.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146174484","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