Biomedical Signal Processing and Control最新文献

{"title":"Style adaptation for avoiding semantic inconsistency in Unsupervised Domain Adaptation medical image segmentation","authors":"Ziqiang Liu ,&nbsp;Zhao-Min Chen ,&nbsp;Huiling Chen ,&nbsp;Shu Teng ,&nbsp;Lei Chen","doi":"10.1016/j.bspc.2025.107573","DOIUrl":"10.1016/j.bspc.2025.107573","url":null,"abstract":"<div><div>As the popularity of Unsupervised Domain Adaptation (UDA) in medical image processing continues to rise, there is a growing interest in utilizing this technology to address data discrepancies arising from multiple centers or devices. Previous methods addressed the UDA by aligning distributions from different domains. However, these methods may suffer from potential semantic inconsistency during domain alignment. To tackle the problem, we introduce a novel UDA framework, named as the Style Adaptation Network (SAN), it achieves domain alignment by separating the style and semantic information and aligning the style information to avoid the issue of semantic inconsistency. Specifically, the SAN contains style extraction and style alignment modules. The style extraction module partition the extracted features along the channel dimension and extracts semantic and style information into designated semantic and style features, respectively. Then, the style alignment module is employed to align the style features from different domains, thereby achieving alignment of data distributions across different domains. We extensively evaluate our approach, including in tasks such as substructure segmentation in the heart and multi-organ segmentation in the abdomen, covering bi-directional cross-modal adaptation, <em>i.e.,</em> MRI and CT. The experimental findings indicate that our SAN can efficiently improve the segmentation performance. In comparison with state-of-the-art approaches, our approach exhibits significant advantages.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"105 ","pages":"Article 107573"},"PeriodicalIF":4.9,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143146780","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":"Symmetric deformable image pairwise registration by optimizing neural fields network architectures for guaranteeing transformations consistency","authors":"Zhenyu Zhu ,&nbsp;Rui Song ,&nbsp;Ying Wei","doi":"10.1016/j.bspc.2024.107453","DOIUrl":"10.1016/j.bspc.2024.107453","url":null,"abstract":"<div><div>In the registration task, the transformation should ideally output a deformation field with diffeomorphic properties, that is, the deformation field is reversible and smooth so that the topology of the image will not change during the transformation. However, the performance of existing registration methods in ensuring the smoothness and reversibility of the deformation field is still unsatisfactory. In this paper, a novel inverse consistency neural field (ICNF) method was proposed, which can guarantee the reversibility of the registration transformation and significantly improve the regularity of the registration deformation field. The proposed method is based on pairwise optimization and takes advantage of the powerful representational capabilities of deep neural networks to model the transformations between the joint symmetric estimation image pairs. Based on the different methods of generating deformation fields, the proposed method can either directly output forward and backward displacement fields, realized as displacement field deformable image registration (<em>ICNF-disp</em>), or generate forward and backward time-dependent velocity fields, and integrate these velocity fields to derive the deformation fields, known as diffeomorphic deformable image registration (<em>ICNF-diff</em>). To reduce the impact of interpolation errors in the generated velocity field on <em>ICNF-diff</em> registration performance, we propose a novel inverse consistent loss function for the velocity field. By imposing an inverse consistency constraint on the time-dependent velocity vector, the invertibility and topological preservation of the transformation are further ensured. Extensive experiments on a public Magnetic Resonance 3D brain scan dataset show that the proposed method guarantees invertibility of the transformation between image pairs while outperforming the state-of-the-art registration methods on registration regularity (<em>ICNF-disp</em> improved 86.03% and <em>ICNF-diff</em> improved 97.12%).</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"104 ","pages":"Article 107453"},"PeriodicalIF":4.9,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143155301","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":"Dynamic residual distillation Cycle-Consistent GAN for unpaired ultrasound image translation","authors":"Cunang Jiang,&nbsp;Yaqi Wang,&nbsp;Dai Yu,&nbsp;Jianxun Zhang","doi":"10.1016/j.bspc.2025.107578","DOIUrl":"10.1016/j.bspc.2025.107578","url":null,"abstract":"<div><div>As an affordable, real-time, and radiation-free medical imaging modality, ultrasound is widely used for organ detection and disease observation. Handheld ultrasound devices, with their convenience and portability, have expanded applications in primary care, bedside monitoring, and disaster relief. However, these devices often produce lower-quality images compared to traditional ultrasound systems, limiting their diagnostic potential. To address this challenge, we propose an improved CycleGAN framework tailored for unpaired ultrasound image enhancement. Specifically, we introduce residual feature distillation blocks (RFDB) to effectively integrate multi-level features within the generator and adopt a dynamic discriminator design that gradually reduces filter numbers during training to improve efficiency. Furthermore, an unbalanced training strategy is employed to mitigate convergence issues. We compare our proposed method with nine image enhancement methods on five organ datasets, including five traditional methods and four learning-based methods. For example, on the Breast dataset, compared to the state-of-the-art traditional methods and learning-based methods, our approach reduces MAE by 4% and 11%, respectively, while improving PSNR by 7% and 3%. Similar improvements are observed across other organ datasets, such as Carotid, Kidney, Liver, and Thyroid, demonstrating the robustness and generalizability of our approach.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"105 ","pages":"Article 107578"},"PeriodicalIF":4.9,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143146817","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":"Understandable time frame-based biosignal processing","authors":"Hamed Rafiei,&nbsp;Mohammad-R. Akbarzadeh-T","doi":"10.1016/j.bspc.2024.107429","DOIUrl":"10.1016/j.bspc.2024.107429","url":null,"abstract":"<div><div>The explainability of biological time series poses considerable challenges regarding signal multiplicity, high volatility, nonstationarity, and noisiness in pursuit of understanding human intentions and conditions. These challenges often arise since data points in the time series are inherently unexplainable and need complex models for proper processing. Here, we propose data frames as a primary information unit. Specifically, the proposed biosignal time frame (BioTF) series incorporates data frames inspired by candlestick components from financial data analysis, such as starting, highest, lowest, and ending values (SHLE). We implement BioTF on four benchmarked biosignal classification tasks, including electromyography (EMG), high-density surface electromyography (HD-sEMG), electroencephalography (EEG), and electrocardiogram (ECG). We study various time frame lengths, components, network activation functions, and architectures for these instances. The bio time-frame representation shows similar patterns, technical analysis, and results to financial data analysis, offering an exciting analogy between these two domains. Compared with several point-based strategies, the proposed BioTF improves temporospatial explainability and achieves as much as 7% improved classification due to the reduced complexity by extracting intuitive features in the proposed frame-based representation. The proposed BioTF furthermore leads to competitive results using simpler networks with as much as four times faster end-to-end training and lighter frame-based feature extraction after-step training. The proposed method enables lightweight and transparent implementations of AI recommendation systems for expert manipulation and trustworthy medical translations of bio time series. The proposed SHLE representation is general and could be extended towards more detailed signal representations.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"103 ","pages":"Article 107429"},"PeriodicalIF":4.9,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143098760","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":"CE-YOLO: A channel-efficient YOLO algorithm for colorectal polyp detection","authors":"Chao Xu ,&nbsp;Zhiwei Fan ,&nbsp;Yaoyao Ma ,&nbsp;Yuling Huang ,&nbsp;Jing Wang ,&nbsp;Yishen Xu ,&nbsp;Di Wu","doi":"10.1016/j.bspc.2025.107529","DOIUrl":"10.1016/j.bspc.2025.107529","url":null,"abstract":"<div><div>Colorectal cancer (CRC) represents a substantial public health challenge, with early detection of precancerous lesions, notably polyps, via colonoscopy being pivotal for timely intervention. Many deep learning-based polyp detection methods have been proposed to improve the polyp detection rate, but these methods are often too complex for clinical edge deployment. In this study, we propose a lightweight, channel-efficient YOLO algorithm, termed CE-YOLO, specifically designed for detecting colorectal polyps. We present the PCST module for feature extraction, develop the DCBM structure for feature fusion and transmission, and create an appropriate detection head. The PCST module combines spatial transformation convolution with a cross-stage partial connection structure and incorporates partial channel convolution. This integration allows for better capture and retention of fine-grained features while significantly reducing computational complexity without compromising detection performance. The DCBM is a multi-branch, multi-scale feature fusion structure utilizing dynamic convolution kernels, facilitating efficient fusion and propagation of multi-scale features along the channel dimension. Additionally, we propose and apply a mixed-precision quantization strategy based on sensitivity traversal analysis, a first in polyp detection, reducing model size for efficient edge deployment while maintaining accuracy. We conduct experiments on our proposed method and existing state-of-the-art object detection algorithms using six datasets, to better evaluate detection performance and generalization capability. The experimental results demonstrate that our model achieves superior performance with the lowest complexity, validating the efficacy and benefits of our approach. Compared to existing studies, our research emphasizes efficiency and lightweight, offering greater potential for clinical application.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"104 ","pages":"Article 107529"},"PeriodicalIF":4.9,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143154738","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 new methodological approach based on stationarity and permutation entropy of EMG Bursts for assessing muscle function alterations in a Parkinson’s disease animal model","authors":"Fernando D. Farfán ,&nbsp;Ana L. Albarracín ,&nbsp;Gabriel A. Ruíz ,&nbsp;Eduardo Fernández-Jover","doi":"10.1016/j.bspc.2025.107638","DOIUrl":"10.1016/j.bspc.2025.107638","url":null,"abstract":"<div><div>The EMG signal is the electrical manifestation of motor unit recruitment processes underlying the contractile dynamics of muscle fibers. The analysis methodology frequently carried out includes a preprocessing stage based on artifact removal, stationarity testing, a feature extraction and interpretation stage. Generally, stationarity criteria are difficult to meet when EMG signals are evoked by momentary activations (bursting activity). Thus, the study of contractile patterns evoked in free-moving protocols require particular treatments. Here, we propose a new approach for quantitatively measuring stationarity by using mean, variance and autocovariance test (MVA-test) and the Permutation Entropy for measuring uncertainty degree. This methodology was applied to EMG signals obtained from a Parkinson’s disease (PD) lesion model to longitudinally study the muscle function alterations. The MVA-test revealed that the temporal structure of EMG around the reference contractile zone presents incremental non-stationary characteristics over post-injury time. Likewise, it was observed that the initial phase of motor recruitment in the biceps femoris (BF) muscle (around the onset) presents a high non-stationary component, which increases over post-injury time. Permutation entropy measures throughout the contractile dynamics of the BF muscle revealed that the uncertainty degree decreases in the initial phase of contraction as the animal’s post-injury time increases. The analysis proposed allowed for a longitudinal characterization of muscle function alterations in an animal model of PD in terms of the stationarity properties of EMG signals. Furthermore, it was observed that permutation entropy could serve as a robust biomarker for quantifying neuromuscular remodeling caused by PD progression.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"104 ","pages":"Article 107638"},"PeriodicalIF":4.9,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143155332","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":"Automated diagnosis of plus disease in retinopathy of prematurity based on transformer-based unsupervised curriculum learning","authors":"K. Deepthi ,&nbsp;M.S. Josephine ,&nbsp;V. Jayabala Raja","doi":"10.1016/j.bspc.2025.107521","DOIUrl":"10.1016/j.bspc.2025.107521","url":null,"abstract":"<div><div>Retinopathy of prematurity (ROP) is a serious ocular condition that predominantly impacts premature infants, with plus disease serving as a significant marker of its severity. Timely and accurate identification of plus disease is essential for effective intervention and the prevention of vision impairment. To overcome this issue, a novel Transformer-based Unsupervised Curriculum Learning model (TUCLM) is introduced for automatic diagnosis of plus disease. The proposed Transformer-based Unsupervised Curriculum Learning Model (TUCLM) is an advanced framework for automating the diagnosis of several diseases without the need for labelled training data. The unsupervised clustering component processes these features to group them into clusters and assigns pseudo-labels, facilitating a pseudo-supervised learning approach. The training method begins with the PRes-SE-att-ViT extracting features from retinopathy images. These features are then examined using an unsupervised clustering method, which categorizes similar features into distinct clusters. To enhance the quality of learning, the nearest neighbour of centroid theory is used to identify reliable samples from each cluster. These selected samples are used to fine-tune the PRes-SE-att-ViT model. After fine-tuning, the model is re-used to extract features from the complete dataset, which includes both reliable samples and additional images. The features are reclustered, pseudo-labels are updated, and reliable samples are reselected using the new clustering results. This iterative cycle of feature extraction, clustering, pseudo-labeling, and fine-tuning is repeated until the PRes-SE-att-ViT model reaches convergence. The experimental results demonstrate good results in detecting plus disease with accuracy of 99.3%, precision of 98.8%, recall of 99.1%, and F1-score of 99%.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"104 ","pages":"Article 107521"},"PeriodicalIF":4.9,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143155300","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":"DBMAE-Net: A dual branch multi-scale feature adaptive extraction network for retinal arteriovenous vessel segmentation","authors":"Cheng Wan ,&nbsp;Jianhong Cheng ,&nbsp;Weihua Yang ,&nbsp;Lu Chen","doi":"10.1016/j.bspc.2025.107619","DOIUrl":"10.1016/j.bspc.2025.107619","url":null,"abstract":"<div><div>Vessel morphology and tube diameter changes in the retina are closely related to diseases such as diabetic retinopathy and glaucoma. Traditional methods of vessel segmentation and tube diameter measurement suffer from insufficient accuracy and low automation. To address these limitations, this paper proposes a retinal arteriovenous segmentation network model based on dual branch multi-scale feature adaptive extraction. The model is designed to tackle the challenge of complex and similar arteriovenous vessel structures by incorporating a multi-scale feature extraction and fusion module, which enhances the capture of contextual vessel information and microvascular features. To ensure the model learns rich semantic information, a dual-branch encoder is implemented to extract multi-scale and residual features. The multi-feature local global attention fusion module then performs feature fusion, integrating these features effectively. In the down-sampling module, an adaptive pooling mechanism is introduced to balance detailed and global information of the arterial vessels. This adaptive learning of hyperparameters helps to mitigate the confusion between arterial and venous vessels. Furthermore, a multi-layer semantic supervision module is incorporated to enhance the model’s learning ability and improve segmentation accuracy for arteriovenous vessel features. The method’s effectiveness is validated on the DRIVE-AV, HRF-AV, and LES-AV datasets, achieving accuracy rates of 94.19%, 99.26%, and 93.10%, respectively. Additionally, an equivalent diameter measurement method based on optic disc and macular localization is proposed. This method accurately calculates the equivalent diameter of arterioles and veins, providing quantitative indices for the auxiliary diagnosis of related diseases and enhancing clinical efficiency.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"104 ","pages":"Article 107619"},"PeriodicalIF":4.9,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143155331","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":"Hybrid EEG-fNIRS decoding with dynamic graph convolutional-capsule networks for motor imagery/execution","authors":"Hongtao Wang ,&nbsp;Zhizheng Yuan ,&nbsp;Haiyan Zhang ,&nbsp;Feng Wan ,&nbsp;Yu Li ,&nbsp;Tao Xu","doi":"10.1016/j.bspc.2025.107570","DOIUrl":"10.1016/j.bspc.2025.107570","url":null,"abstract":"<div><div>In this study, we proposed a cascade structure of dynamic graph convolutional and capsule networks for accurate decoding of motor imagery (MI) based brain-computer interfaces (BCIs) with both electroencephalogram signals and functional near-infrared spectroscopy (fNIRS) signals. The same network structure with different parameter settings was applied to these two modalities to extract features through temporal convolution block, dynamic graph convolution block, and capsule generation block. The temporal convolution block was used to learn temporal features, the dynamic graph convolution block to learn spatial features, and the capsule generation block to generate primary capsules. Then the capsuled features will undergo cross-attention and then go through a feature fusion block and a dynamic routing block which is an iterative algorithm designed to learn the connection weights between primary capsules and digit capsules. The mean accuracy of leave-one-session-out testing can reach 92.60 %±4.49 % and 92.20 %±2.95 % for self-collected EEG-fNIRS data (dataset A) and publicly available dataset (dataset B) whereas the accuracy of randomized five-fold cross-validation testing for another publicly available dataset (dataset C) is 85.30 %±3.58 %. Moreover, the leave-one-subject-out testing shows that the proposed method is superior to that of the current state-of-the-art methods, like hybrid EEGNet, hybrid LSTM, or hybrid CapsNet at least 4 % across all three datasets. These results demonstrate that the proposed network structure can be a good candidate for the decoding of MI-based BCIs with multiple modalities.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"104 ","pages":"Article 107570"},"PeriodicalIF":4.9,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143155341","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":"MAFBN: Multi-Attention Forward and Backward network to predict epileptic seizure","authors":"Tawfeeq Shawly ,&nbsp;Ahmed A. Alsheikhy ,&nbsp;Yahia Said ,&nbsp;Aws I. AbuEid ,&nbsp;Abdulrahman A. Alzahrani ,&nbsp;Abdulrahman A. Alshdadi ,&nbsp;Hossam E. Ahmed","doi":"10.1016/j.bspc.2025.107574","DOIUrl":"10.1016/j.bspc.2025.107574","url":null,"abstract":"<div><div>Epilepsy occurs due to an uncontrolled or unprovoked electrical charge in the brain, which affects functionality and causes people to lose consciousness or stare blankly for a while. Numerous studies and research were conducted to detect this disease and aid neurologists in their evaluations and assessments. Currently, a few studies to identify epilepsy based on deep learning using an electroencephalogram (EEG) have been carried out to extract the right features that are believed to be related to this disease to provide an accurate diagnosis. This article presents a new novel deep-learning-based solution to predict epileptic seizures using EEG. This solution uses a Multi-Attention Forward and Backward Network (MAFBN). It incorporates multi-attention methodologies, forward and backward feedback topology, and U-shaped architecture into one solid structure to identify the disease and send early alerts to healthcare providers. MAFBN achieves noticeable accuracy at nearly 97.88%, while other considered performance indicators varied between 95% and 98.01%. These outcomes show that MAFBN surpasses other developed models and can be implemented in healthcare facilities to provide aid, support, and assistance for physicians and neurologists in their diagnosis.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"104 ","pages":"Article 107574"},"PeriodicalIF":4.9,"publicationDate":"2025-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143155294","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}