Biomedical Signal Processing and Control最新文献

{"title":"An integrated convolutional neural network with zero-dimensional cardiovascular hemodynamics parameters for early cardiovascular disease detection","authors":"Denesh Sooriamoorthy ,&nbsp;Mohammed Ayoub Juman ,&nbsp;Aaruththiran Manoharan ,&nbsp;Marwan Nafea ,&nbsp;Anandan Shanmugam","doi":"10.1016/j.bspc.2025.108171","DOIUrl":"10.1016/j.bspc.2025.108171","url":null,"abstract":"<div><div>This study addresses a critical challenge in cardiovascular disease (CVD) management: late detection, which, often at an advanced stage, can lead to high mortality risk. Conventional approaches to severe CVD cases involve invasive treatments, which can distress patients. To mitigate risk of severe outcomes or sudden death from CVD, this research introduces a novel predictor framework, combining upstream blood pressure waveform analysis with artificial intelligence, specifically integrating a Convolutional Neural Network (CNN) and Rideout’s zero-dimensional cardiovascular model parameters. Rideout’s model identified 16 significant parameters affecting the aortic wave, which were used to train CNN for predicting CVD from aortic waveforms, derived from radial pulse waveforms using two transfer functions. The study pinpointed two critical parameters, Pulmonary Vein 2 and Systemic Aortic Artery 1, as CVD indicators, proposing a biological correlation where these parameters concurrently relax to facilitate smooth blood flow, thereby reducing blood vessels’ resistance values. Experimental validation involved using the best-performing CNN to obtain parameter values from signals in the PhysioNet MIMIC II database, which included 4 CVD and 19 non-CVD signals, serving as base indicators for classifying cardiovascular and non-cardiovascular diseases. These indicators were then used to verify the classification of 3365 healthy signals from the HaeMod dataset and 40 CVD signals collected from Hospital Sultanah Bahiyah (HSB), Malaysia. The system achieved 80.0% and 82.5% accuracy for EIF and GTF transfer functions respectively, based on HSB data, significantly enhancing early detection and offering timely intervention, while proving the potential for practical application of the system in clinical settings.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108171"},"PeriodicalIF":4.9,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144168037","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":"CNN with double-side weighted visibility graph for automated classification of high-frequency oscillations in epilepsy","authors":"Guosheng Yi ,&nbsp;Jingjing Song ,&nbsp;Wenpu Zhang ,&nbsp;Jiang Wang ,&nbsp;Shanshan Li ,&nbsp;Lihui Cai","doi":"10.1016/j.bspc.2025.108178","DOIUrl":"10.1016/j.bspc.2025.108178","url":null,"abstract":"<div><div>High-frequency oscillations (HFOs) in intracranial EEG are deemed clinically significant biomarkers for localization of epileptogenic focus. However, the identification of HFOs historically relies on visual observation which is laborious and invariably prone to inaccuracy. Here, we provided a new strategy that combines double-side weighted visibility graph (dWVG) and convolutional neural network (CNN) to automatically detect HFOs. The effectiveness of the proposed method was confirmed using stereotactic electroencephalography (SEEG) data recorded from 4 epilepsy patients, comprising 1880 HFO fragments and 1880 non-HFO fragments. In addition, the dWVG-based approach was also compared with other graph-based (i.e., time–frequency graph) methods. Results showed that the dWVG approach outperformed the time–frequency diagram and weighted visibility graph (WVG) with better classification accuracy a computing speed. By combining dWVG and CNN based on signal features instead of LeNet-like CNN, the best classification performance was obtained with an accuracy of 95.52%. These results indicate that the proposed approach plays a significant role in HFO classification, which may facilitate the localization of epileptogenic focus and provide more effective treatment for patients.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108178"},"PeriodicalIF":4.9,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144166303","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":"BRMSA-Net: Disclosing concealed colorectal polyps in colonoscopy images via Boundary Recalibration and Multi-Scale Aggregation Network","authors":"Meriem Sebai ,&nbsp;Evgin Goceri","doi":"10.1016/j.bspc.2025.108083","DOIUrl":"10.1016/j.bspc.2025.108083","url":null,"abstract":"<div><div>For the early detection of colorectal cancer, automated polyp segmentation from colonoscopy images is very crucial. Polyps, often amorphous, exhibit a wide range of sizes and shapes. Furthermore, the boundaries separating polyps from the surrounding mucosal tissues are frequently not distinct. Consequently, segmenting polyps in practical settings is a particularly challenging task. In this study, we propose a Boundary Recalibration and Multi-Scale Aggregation Network (BRMSA-Net) for precise and consistent colon polyp segmentation. In particular, we first introduce the Multi-level Feature Aggregation (MFA) decoder to generate a global segmentation map by fusing high-level features for polyp localization. The MFA decoder incorporates the Criss-cross Feature Fusion (CFF) module, which effectively combines the cross-level features and addresses the issue of semantic gap across different levels. The segmentation map predicted by the MFA decoder is improved by the novel Incremental Boundary Refinement (IBR) module. The IBR module includes the Incremental Scale Fusion (ISF) module, which uses dilated convolutions to extract features with different receptive fields and strategically fuses them to reduce feature disparity. These features are then utilized by the Boundary Recalibration Cross-attention (BRC) module, which gradually refines the polyp boundaries by combining spatial and channel cross-attention. Extensive experiments on public polyp datasets show that BRMSA-Net outperforms several competing approaches in learning and generalization. It achieves 94.02% DSC and 89.46% IoU on CVC-ClinicDB, 92.26% DSC and 87.28% IoU on KvasirSEG, 82.45% DSC and 74.52% IoU on CVC-ColonDB, and 82.28% DSC and 73.81% IoU on ETIS-Larib. The code is available at: <span><span>https://github.com/Meriem-Sebai/BRMSA-Net</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108083"},"PeriodicalIF":4.9,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144168032","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":"CPIA dataset: a large-scale comprehensive pathological image analysis dataset for self-supervised learning pre-training","authors":"Nan Ying ,&nbsp;Yanli Lei ,&nbsp;Tianyi Zhang ,&nbsp;Shangqing Lyu ,&nbsp;Sicheng Chen ,&nbsp;Zeyu Liu ,&nbsp;Yunlu Feng ,&nbsp;Yu Zhao ,&nbsp;Guanglei Zhang","doi":"10.1016/j.bspc.2025.108148","DOIUrl":"10.1016/j.bspc.2025.108148","url":null,"abstract":"<div><div>Pathological image analysis is a crucial field in computer-aided diagnosis. Transfer learning using models initialized on natural images has improved the downstream pathological performance. However, the lack of sophisticated domain-specific pathological initialization hinders their potential. Self-supervised learning (SSL) enables pre-training without sample-level labels, overcoming the challenge of expensive annotations. Thus, this field calls for a comprehensive dataset, similar to the ImageNet in computer vision. This work introduces a large-scale comprehensive pathological image analysis (CPIA) dataset for SSL pre-training. The CPIA dataset contains 148,962,586 images, covering over 48 organs/tissues and approximately 100 kinds of diseases, which includes two main data types: whole slide images (WSIs) and regions of interest (ROIs) images. Furthermore, we establish a standard multi-scale pathological data processing workflow, combined with the diagnosis habits of senior pathologists. The CPIA dataset facilitates a comprehensive pathological understanding and enables pattern discovery explorations. Additionally, to launch the CPIA dataset, several state-of-the-art (SOTA) baselines of SSL pre-training and downstream evaluation are specially conducted. The CPIA dataset information and code are available at <span><span>https://github.com/zhanglab2021/CPIA_Dataset</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108148"},"PeriodicalIF":4.9,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144166305","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":"Modeling the neurodegenerative Parkinson’s disease treatment with high effectiveness of beta-band neural activity using the optogenetic-electrical hybrid stimulation","authors":"Shabnam Andalibi Miandoab ,&nbsp;Nazlar Ghasemzadeh","doi":"10.1016/j.bspc.2025.107999","DOIUrl":"10.1016/j.bspc.2025.107999","url":null,"abstract":"<div><div>Optogenetic-electrical hybrid stimulation provides powerful and subtle control over activity of distinctive neurons. Since Basal Ganglia (BG) and Thalamus (Th) of brain are target of damage or stimulation in Parkinson’s disease (PD), we have considered a complete BG-Th network model which consists of all neuronal parts of BG and Th. In this paper, to control signs of PD, we have introduced two hybrid stimulations with different characteristics in considered model, and have investigated the influence process and the function of the neural activities of different parts of brain. To achieve this, hybrid stimulations have been considered in various combinations of two optical stimulations with Halorhodopsin (NpHR) and Channelrhodopsin-2 (ChR2) opsins and monophasic and biphasic electrical deep brain stimulations (DBS). Initially, function of the hybrid stimulation of optogenetic NpHR-Monophasic DBS and ChR2-Biphasic DBS on the abnormal firing pattern and synchronized excessive oscillations of the beta activity have been examined. Error Index (EI), beta band activity <span><math><mrow><mo>(</mo><msub><mi>β</mi><mrow><mi>a</mi><mi>c</mi><mi>t</mi></mrow></msub><mo>)</mo></mrow></math></span> and average firing rate (AFR) and Local Field Potential (LFP) are the evaluation criteria that we have used to effectiveness of our proposed method on neural activity of BG-Th model. According to the results, hybrid stimulation of NpHR optical stimulation beside mono phase electrical stimulation with effective results has been introduced as optimal stimulation. Our findings reveal that hybrid stimulation in low intensity and duration has the most influence on suppression of the abnormal PD behaviour and causes the least tissue damage. Besides, our proposed method supplies novel insights for accurate control on PD.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 107999"},"PeriodicalIF":4.9,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144168034","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":"Integrating intratumor-peritumor implicit correlation and deep semantic features for ultrasound breast cancer diagnosis","authors":"Yuan Zheng ,&nbsp;Shujun Xia ,&nbsp;Zhao Yao ,&nbsp;Jianqiao Zhou ,&nbsp;Jinhua Yu ,&nbsp;Guoqing Wu","doi":"10.1016/j.bspc.2025.108017","DOIUrl":"10.1016/j.bspc.2025.108017","url":null,"abstract":"<div><h3>Purpose</h3><div>Breast cancer remains a significant cause of cancer-related mortality among women globally, prompting the development of intelligent diagnostic tools based on ultrasound images. In this study, an effective multi-region spatial correlation mining method was proposed to enhance breast cancer diagnostic accuracy.</div></div><div><h3>Methods</h3><div>Denoting the tumor core and surrounding peritumoral regions as nodes, a Graph Convolutional Network (GCN)-based model is developed to capture multi-region correlations between adjacent nodes. Considering the limitations of GCN in extracting fine-grained image details, a medical foundation model is introduced to extract generic semantic features from tumor images. Additionally, a cross-attention module is designed to integrate the global multi-region correlation and high-level semantic features. Finally, the proposed dual-branch model is employed for the benign and malignant classification and lymph node metastasis prediction of breast cancer.</div></div><div><h3>Results</h3><div>Extensive comparative and ablation experiments were conducted on 4,053 cases from two independent datasets. For breast cancer diagnosis, our method achieved superior performance with accuracy improvements of 1.85 % and 2.34 % over the second-best results in two datasets respectively, reaching 97.40 % (±2.579) and 88.98 % (±1.108). In lymph node metastasis prediction, our model demonstrated a 3.07 % accuracy enhancement, reaching 71.11 % (±1.490). Visualization analyses using Class Activation Mapping (CAM) and Integrated Gradients (IG) further validated the model’s effectiveness in capturing discriminative features in intratumor-peritumor regions.</div></div><div><h3>Conclusion</h3><div>Our method demonstrates superior performance in both breast tumor classification and lymph node metastasis prediction, with its effectiveness attributed to the synergistic integration of multi-region correlations and deep semantic features.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108017"},"PeriodicalIF":4.9,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144166878","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":"VesselMamba: 3D vessel segmentation in CTA images using Mamba with enhanced Spatial-Channel Attention","authors":"Ziyue Xie ,&nbsp;Xiaoquan Huang ,&nbsp;Shiyao Chen ,&nbsp;Yonghong Shi","doi":"10.1016/j.bspc.2025.107982","DOIUrl":"10.1016/j.bspc.2025.107982","url":null,"abstract":"<div><div>3D vessel segmentation in Computed Tomography Angiography (CTA) is crucial yet challenging due to the complex, multi-scale, and elongated branching structure of human vasculature. Accurate modeling requires capturing both long-range dependencies and multi-scale information inherent in vascular networks. While deep neural networks like CNNs and Vision Transformers (ViTs) have demonstrated progress, they often face challenges balancing global receptive field capture with computational efficiency. To address these limitations, we propose VesselMamba, a novel 3D vessel segmentation framework based on Mamba, an approach for modeling long-range dependencies with linear complexity. VesselMamba integrates parallel Mamba blocks in the encoder to efficiently capture vascular continuity and long-range dependencies. Additionally, the encoder is enhanced with a Spatial-Channel Attention with Spatial Pyramid Pooling (SCASPP) module to effectively model multi-scale information and optimize the integration of global and local features, significantly improving segmentation precision. Furthermore, a composite loss function that combines the clDice loss with traditional cross-entropy and Dice losses is employed to improve the connectivity of segmented vessels. This reduces fragmentation and artifacts, leading to more reliable segmentations. Comprehensive ablation studies on private and public datasets demonstrate the complementary nature and effectiveness of the proposed modules. Experimental results show that VesselMamba achieves state-of-the-art performance in CTA vessel segmentation tasks, outperforming existing methods and providing a robust tool for clinical diagnosis and research.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 107982"},"PeriodicalIF":4.9,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144147244","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":"Interpretability and accuracy of machine learning algorithms for biomedical time series analysis – a scoping review","authors":"Alan Jovic,&nbsp;Nikolina Frid,&nbsp;Karla Brkic,&nbsp;Mario Cifrek","doi":"10.1016/j.bspc.2025.108153","DOIUrl":"10.1016/j.bspc.2025.108153","url":null,"abstract":"<div><div>Current research in biomedical time series (BTS) (e.g., ECG, EEG) analysis focuses on applications of various deep learning approaches to improve classification, prediction, or assessment of states and disorders. When trained on sufficiently large datasets, such approaches mostly lead to highly accurate, yet uninterpretable models, sometimes with a possibility for post-hoc explainability. Since high-stake areas such as healthcare warrant model explanations and, where possible, high interpretability in addition to model efficiency, there is nowadays a surprising scarcity of interpretable machine learning models proposed in this field. Although the machine learning community is aware of the need for interpretable machine learning in BTS analysis, the proposed models do not reflect this need. In this scoping review, we considered over 30,000 studies from the Web of Science database, screened nearly 500 studies, and selected over 50 high-quality studies for detailed analysis. These studies focus on interpretable methods, accurate methods, and approaches bridging the two. Most studies analyzed ECG and EEG signals and concentrated on a limited range of applications, including emotion recognition, heart diseases, epilepsy, and motor imagery, reflecting the scarcity of quality public datasets. K-nearest neighbors and decision trees were the most used interpretable methods, while convolutional neural networks with recurrent or attention layers, achieved the highest accuracy. The methods that balance interpretability and accuracy in BTS analysis include advanced generalized additive models and optimization-based approaches for decision trees, rule learning, and linear models. These approaches warrant further studies, as only a few of them were applied in BTS analysis.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108153"},"PeriodicalIF":4.9,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144166302","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":"Kalman filter in quality assurance in radiotherapy: A practical application for daily dose quality control","authors":"Dimitri Reynard ,&nbsp;Jean-Baptiste Billet ,&nbsp;Alain Barraud ,&nbsp;Christophe Mazzara","doi":"10.1016/j.bspc.2025.108143","DOIUrl":"10.1016/j.bspc.2025.108143","url":null,"abstract":"<div><div>This study introduces a Kalman filter (KF)-based approach to enhance the daily dose quality control of radiotherapy equipment. The radiation production system is modeled as a dynamic system governed by state equations. The KF is applied to 30 months of daily dose quality control (DDQC) data from Varian® Halcyon systems delivering 6 MV flattening filter-free beams. Input measurements for the KF derive from quality control data collected with the Sun Nuclear® DailyQA 3 detector. Monitor units, rather than time, serve as the independent variable, with two iteration frequencies evaluated. The evolution model includes terms for monitor chamber aging and sensitivity corrections based on atmospheric pressure, with a second model further accounting for room temperature.</div><div>The KF extracts additional insights from the quality control measurements. Outliers are detected using a two-standard-deviation window, and drift prediction enables proactive scheduling of dose recalibrations. A significant correlation is observed between KF outputs and machine interventions, such as maintenance, recalibration, and component replacement.</div><div>Further refinements in the evolution model and the inclusion of additional input measurements could improve precision. The systematic collection and automated analysis of machine event logs could also enhance early issue detection and provide a more robust framework for decision-making. The lightweight and computationally efficient nature of KF models, combined with their scalability, suggests they could become a valuable tool for establishing a proactive, data-driven paradigm in radiotherapy quality assurance.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108143"},"PeriodicalIF":4.9,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144147242","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":"Physiologic control of rotary blood pumps by ventricular chamber size estimation using resonantly coupled sensors","authors":"Yu Wang ,&nbsp;George A.F. Ghali ,&nbsp;Xiaoyue Luo ,&nbsp;Pramod Bonde ,&nbsp;Guruprasad A. Giridharan","doi":"10.1016/j.bspc.2025.108135","DOIUrl":"10.1016/j.bspc.2025.108135","url":null,"abstract":"<div><div>Rotary blood pumps (RBP) currently operate at a fixed pump speed and are unable to meet physiologic demand and susceptible to ventricular suction. To overcome this limitation, we developed a left ventricular end-diastolic volume (EDV) based physiologic control algorithm using resonantly coupled high-efficiency sensors. The resonantly coupled sensors consist of apical and outflow sensors that can accurately assess the ventricular chamber size with minimal long-term drift (∼1 %) at 9 months. The ability of the control algorithm was evaluated using an in-silico circulatory system model coupled to an axial or centrifugal flow RBP with 15 % uniformly distributed measurement noise. The EDV setpoint was set to 85 ml, and the efficacy of the EDV control algorithm was evaluated and compared to maintaining a constant pump speed during (1) rest and exercise; (2) rapid, eight-fold augmentation of pulmonary vascular resistance; and (3) rapid transitions between rest and exercise. Safety and robustness of the algorithm was also evaluated by assuming a 6 % volume drift. The EDV control algorithm provided sufficient physiological perfusion and avoided ventricular suction in all cases. Performance of the EDV algorithm was superior compared to maintaining constant pump speed for both types of RBP, demonstrating pump independence of the proposed algorithm.</div></div>","PeriodicalId":55362,"journal":{"name":"Biomedical Signal Processing and Control","volume":"110 ","pages":"Article 108135"},"PeriodicalIF":4.9,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144147455","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}