Computers in biology and medicine最新文献

{"title":"Modeling and simulation of cardiovascular system under cardiac arrest for finding a more effective CPR technique","authors":"Ali Raza ,&nbsp;Hassan Sultan ,&nbsp;Syed Muhammad Abdul Rehman ,&nbsp;Rashid Mazhar ,&nbsp;Tahir Hamid","doi":"10.1016/j.compbiomed.2025.109890","DOIUrl":"10.1016/j.compbiomed.2025.109890","url":null,"abstract":"<div><div>Cardio-Pulmonary Resuscitation (CPR) saves life. However, all the current CPR methods produce only one third to one quarter of the normal cardiac output and hence post-CPR survival has remained very poor. We report a better CPR technique exhibiting increased cardiac output as compared to existing techniques. Obviously, one cannot perform such studies on humans; therefore, we developed a fluidic model of the cardiovascular system under cardiac arrest. This enabled us to actuate different organs independently, sequentially and/or combinatorially to find the most effective CPR technique. Extensive simulations were performed using Simscape®. Our novel combination (combination-1) shows 10.75% improvement in peak aortic pressure and 8.3% improvement in peak cardiac flow-rate with 120 compressions per minute with respect to the baseline CPR method as per AHA/ERC guidelines. Similar improvements were observed at compression rates of 80 and 100 per minute. In addition to finding a more effective CPR technique, we also present our passive cardiovascular model as an open-source software package where different preconditions and modalities can be set prior to conducting the cardiovascular simulations. Thus, it may also serve as a simulator to explore the cardiovascular system behaviors as well as the effects of different contributing factors.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"189 ","pages":"Article 109890"},"PeriodicalIF":7.0,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143488696","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":"Deep learning-based LDL-C level prediction and explainable AI interpretation","authors":"Ali Öter","doi":"10.1016/j.compbiomed.2025.109905","DOIUrl":"10.1016/j.compbiomed.2025.109905","url":null,"abstract":"<div><div>This study investigates the use of deep learning (DL) models to predict low-density lipoprotein cholesterol (LDL-C) levels. The dataset obtained from New York-Presbyterian Hospital/Weill Cornell Medical Center includes triglycerides (TG), total cholesterol (TC) and high-density lipoprotein cholesterol (HDL-C). LDL-C prediction was performed using DL models such as CNN, RNN and LSTM and the results were compared with traditional machine learning (ML) and LDL-C formulas. The obtained results showed that DL models are more successful than traditional formulas while giving closer results to ML models. It is shown that DL models can predict LDL-C with higher accuracy compared to the Sampson, and Martin equation. In particular, RNN and LSTM models performed better in LDL-C prediction than the other formulas. In addition, the prediction results of DL models were explained using Local Interpretable Model-Agnostic Explanations (LIME) method. The features of the proposed models provide more parameters to explain the AI Model better in comparison with the ML models but require more computational efforts to explain DL model decisions. The results demonstrate that DL models in predicting LDL-C levels are more effective than traditional methods for LDL-C prediction and can be used in clinical applications. As a result, the findings might provide significant contributions to assessing cardiovascular disease risk and planning treatment protocols.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"188 ","pages":"Article 109905"},"PeriodicalIF":7.0,"publicationDate":"2025-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143487273","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":"Computational fluid dynamics model utilizing proper orthogonal decomposition to assess coronary physiology and wall shear stress","authors":"Amir Lotfi ,&nbsp;Daniela Caraeni ,&nbsp;Omar Haider ,&nbsp;Abdullah Pervaiz ,&nbsp;Yahya Modarres-Sadeghi","doi":"10.1016/j.compbiomed.2025.109840","DOIUrl":"10.1016/j.compbiomed.2025.109840","url":null,"abstract":"<div><h3>Background</h3><div>Percutaneous coronary intervention (PCI) to alleviate symptoms and improve outcomes in patients with symptomatic coronary artery disease. However, conventional assessments like coronary angiography may not fully capture the hemodynamic significance of coronary lesions. This study explores the utility of Proper Orthogonal Decomposition (POD) in elucidating coronary flow dynamics pre- and post-stent placement.</div></div><div><h3>Objectives</h3><div>Through the utilization of POD modes, we aim to analyze the intricate geometries of individual patients, extracting dominant POD modes both pre- and post-PCI. By engaging these modes, our objective is to discern changes in velocity patterns and wall shear stress, offering insight into the physiological outcomes of stent interventions in coronary arteries.</div></div><div><h3>Methods</h3><div>The POD method with QR-decomposition was employed to generate POD modes, decomposing the vector field of interest into spatial functions modulated by time coefficients. Patients with prior coronary artery bypass surgery, myocardial bridging, collateral arteries, or recent myocardial infarction within 48 h were excluded from the study.</div></div><div><h3>Results</h3><div>Results demonstrated improved hemodynamic parameters post-PCI, with significant enhancements in coronary flow reserve and reduced wall shear stress. POD analysis revealed that the first five modes effectively characterized flow features, highlighting stenosis, stent deployment, and branch dynamics.</div></div><div><h3>Conclusion</h3><div>This exploratory study demonstrates POD's potential for real-time assessment of coronary lesion significance and post-intervention outcomes. Its efficiency in capturing key flow characteristics offers a promising tool for personalized decision-making in interventional cardiology, enhancing our understanding of coronary hemodynamics and optimizing treatment strategies.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"188 ","pages":"Article 109840"},"PeriodicalIF":7.0,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143479965","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":"Advancing personalized immunotherapy for melanoma: Integrating immunoinformatics in multi-epitope vaccine development, neoantigen identification via NGS, and immune simulation evaluation","authors":"Mohammad Javad Kamali ,&nbsp;Mohammad Salehi ,&nbsp;Mohsen Karami Fath","doi":"10.1016/j.compbiomed.2025.109885","DOIUrl":"10.1016/j.compbiomed.2025.109885","url":null,"abstract":"<div><div>The use of cancer vaccines represents a promising avenue in cancer immunotherapy. Advances in next-generation sequencing (NGS) technology, coupled with the development of sophisticated analysis tools, have enabled the identification of somatic mutations by comparing genetic sequences between normal and tumor samples. Tumor neoantigens, derived from these mutations, have emerged as potential candidates for therapeutic cancer vaccines. In this study, raw NGS data from two melanoma patients (NCI_3903 and NCI_3998) were analyzed using publicly available SRA datasets from NCBI to identify patient-specific neoantigens. A comprehensive pipeline was employed to select candidate peptides based on their antigenicity, immunogenicity, physicochemical properties, and toxicity profiles. These validated epitopes were utilized to design multi-epitope chimeric vaccines tailored to each patient. Peptide linkers were employed to connect the epitopes, ensuring optimal vaccine structure and function. The two-dimensional (2D) and three-dimensional (3D) structures of the chimeric vaccines were predicted and refined to ensure structural stability and immunogenicity. Furthermore, molecular docking simulations were conducted to evaluate the binding interactions between the vaccine chimeras and the HLA class I receptors, confirming their potential to elicit a robust immune response. This work highlights a personalized approach to cancer vaccine development, demonstrating the feasibility of utilizing neoantigen-based immunoinformatics pipelines to design patient-specific therapeutic vaccines for melanoma.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"188 ","pages":"Article 109885"},"PeriodicalIF":7.0,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143480069","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":"Corrigendum to “Topological radiogenomics based on persistent lifetime images for identification of epidermal growth factor receptor mutation in patients with non-small cell lung tumors” [Computers in Biology and Medicine. 185 (2025) 109519]","authors":"Takumi Kodama ,&nbsp;Hidetaka Arimura ,&nbsp;Tomoki Tokuda ,&nbsp;Kentaro Tanaka ,&nbsp;Hidetake Yabuuchi ,&nbsp;Nadia Fareeda Muhammad Gowdh ,&nbsp;Chong-Kin Liam ,&nbsp;Chee-Shee Chai ,&nbsp;Kwan Hoong Ng","doi":"10.1016/j.compbiomed.2025.109860","DOIUrl":"10.1016/j.compbiomed.2025.109860","url":null,"abstract":"","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"188 ","pages":"Article 109860"},"PeriodicalIF":7.0,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143499554","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Intelligent larval zebrafish phenotype recognition via attention mechanism for high-throughput screening","authors":"Baihua Wang ,&nbsp;Qi Sun ,&nbsp;Yujia Liu ,&nbsp;Jiheng Zhang ,&nbsp;Gaozheng Li ,&nbsp;Sifang Wu ,&nbsp;Houbing Zheng ,&nbsp;Jialin Ye ,&nbsp;Meihua Zhou ,&nbsp;Haisu Zheng ,&nbsp;Yongqiang Yu ,&nbsp;Yi Zhong ,&nbsp;Yuanzi Wu ,&nbsp;Da Huang ,&nbsp;Biao Wang ,&nbsp;Zuquan Weng","doi":"10.1016/j.compbiomed.2025.109892","DOIUrl":"10.1016/j.compbiomed.2025.109892","url":null,"abstract":"<div><h3>Background</h3><div>Larval zebrafish phenotypes serve as critical research indicators in fields such as ecotoxicology and safety assessment since phenotypic defects are closely related to alterations of underlying pathway. However, identifying these defects is time-consuming and requires specialized knowledge.</div></div><div><h3>Method</h3><div>We proposed a deep network model called RECNet, which combines attention mechanisms and residual structures. In terms of data processing, we applied the mixup data augmentation technique and accumulated a collection of 6805 larval zebrafish phenotype images, mostly generated from our laboratory. Our proposed model was deployed to execute two distinct tasks, including a four-classification of zebrafish phenotypes and a seven-classification involving mixed labels for abnormalities.</div></div><div><h3>Results</h3><div>In the four-class classification task, the RECNet model achieved an accuracy of 0.949, with a mean area under the curve of 0.986 and an F1-score of 0.966. Through interpretable research, attention mechanisms enable the model to focus more accurately on regions of interest. In the mixed-label seven-classification task for anomalies, our model achieved an accuracy of 0.913 and a mean average precision value of 0.847 by employing the weighted loss function (DFBLoss). Furthermore, in a new test dataset, the RECNet model achieved accuracy rates of 0.924 and 0.876 for the two tasks, respectively. Our RECNet model was trained by orders of magnitude larger dataset than previous studies and also showed better accuracy rates.</div></div><div><h3>Conclusions</h3><div>Our method holds promise for diverse applications within zebrafish laboratories and fields such as toxicology, providing indispensable support to scientific research.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"188 ","pages":"Article 109892"},"PeriodicalIF":7.0,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143479963","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":"CephTransXnet: An attention enhanced feature fusion network leveraging neighborhood rough set approach for cephalometric landmark prediction","authors":"R. Neeraja,&nbsp;L. Jani Anbarasi","doi":"10.1016/j.compbiomed.2025.109891","DOIUrl":"10.1016/j.compbiomed.2025.109891","url":null,"abstract":"&lt;div&gt;&lt;div&gt;The convergence of medical imaging, computer vision, and orthodontics has made automatic cephalometric landmark detection a pivotal area of research. Accurate cephalometric analysis is crucial in orthodontics, orthognathic and maxillofacial surgery for diagnosis, treatment planning, and monitoring craniofacial growth. In this research study, a multi-branch fused feature extraction network titled &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;C&lt;/mi&gt;&lt;mi&gt;e&lt;/mi&gt;&lt;mi&gt;p&lt;/mi&gt;&lt;mi&gt;h&lt;/mi&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;mi&gt;r&lt;/mi&gt;&lt;mi&gt;a&lt;/mi&gt;&lt;mi&gt;n&lt;/mi&gt;&lt;mi&gt;s&lt;/mi&gt;&lt;mi&gt;X&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mi&gt;n&lt;/mi&gt;&lt;mi&gt;e&lt;/mi&gt;&lt;mi&gt;t&lt;/mi&gt;&lt;/mrow&gt;&lt;/msub&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; is proposed to automatically predict landmark coordinates from cephalometric radiographs. The initial sequential branch enhances discriminative local feature learning and feature extraction through parallel feature fusion by integrating Convolved Pooled Normalized (&lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;C&lt;/mi&gt;&lt;mi&gt;P&lt;/mi&gt;&lt;mi&gt;N&lt;/mi&gt;&lt;/mrow&gt;&lt;mi&gt;B&lt;/mi&gt;&lt;/msub&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;) and Gradient Optimized Multi-Path Bottleneck (&lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;G&lt;/mi&gt;&lt;mi&gt;M&lt;/mi&gt;&lt;mi&gt;B&lt;/mi&gt;&lt;/mrow&gt;&lt;mi&gt;B&lt;/mi&gt;&lt;/msub&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;) blocks with Channel and Spatial Attention (&lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;C&lt;/mi&gt;&lt;mi&gt;S&lt;/mi&gt;&lt;mi&gt;A&lt;/mi&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;/mrow&gt;&lt;mi&gt;M&lt;/mi&gt;&lt;/msub&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;) module. The Swin Transformer (&lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;S&lt;/mi&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;/mrow&gt;&lt;mi&gt;B&lt;/mi&gt;&lt;/msub&gt;&lt;mo&gt;)&lt;/mo&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; branch efficiently handles long-range dependencies and extracts global features in cephalometric radiographs. The multi-branch fused features along with features from skip connections of &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;C&lt;/mi&gt;&lt;mi&gt;P&lt;/mi&gt;&lt;mi&gt;N&lt;/mi&gt;&lt;/mrow&gt;&lt;mi&gt;B&lt;/mi&gt;&lt;/msub&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; and &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;G&lt;/mi&gt;&lt;mi&gt;M&lt;/mi&gt;&lt;mi&gt;B&lt;/mi&gt;&lt;/mrow&gt;&lt;mi&gt;B&lt;/mi&gt;&lt;/msub&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; blocks are concatenated using a Coordinate Attention module &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mo&gt;(&lt;/mo&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;C&lt;/mi&gt;&lt;mi&gt;o&lt;/mi&gt;&lt;mi&gt;A&lt;/mi&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;/mrow&gt;&lt;mi&gt;M&lt;/mi&gt;&lt;/msub&gt;&lt;mo&gt;)&lt;/mo&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; to captures the positional relationships between various landmark features. A Landmark Discriminative Deviation Factor &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mo&gt;(&lt;/mo&gt;&lt;mrow&gt;&lt;mi&gt;L&lt;/mi&gt;&lt;mi&gt;D&lt;/mi&gt;&lt;mi&gt;D&lt;/mi&gt;&lt;mi&gt;F&lt;/mi&gt;&lt;/mrow&gt;&lt;mo&gt;)&lt;/mo&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; is determined by applying the Neighborhood Rough Set &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mo&gt;(&lt;/mo&gt;&lt;mrow&gt;&lt;mi&gt;N&lt;/mi&gt;&lt;mi&gt;R&lt;/mi&gt;&lt;mi&gt;S&lt;/mi&gt;&lt;/mrow&gt;&lt;mo&gt;)&lt;/mo&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; approach to analyse the surrounding features of each landmark by considering spatial relationships or similarity measures between the landmarks and neighboring regions. The Spatial Pyramid Pooling (&lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;S&lt;/mi&gt;&lt;mi&gt;P&lt;/mi&gt;&lt;mi&gt;P&lt;/mi&gt;&lt;/mrow&gt;&lt;mi&gt;L&lt;/mi&gt;&lt;/msub&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;) layer incorporated in the final phase of &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;msub&gt;&lt;mrow&gt;&lt;mi&gt;C&lt;/mi&gt;&lt;mi&gt;e&lt;/mi&gt;&lt;mi&gt;p&lt;/mi&gt;&lt;mi&gt;h&lt;/mi&gt;&lt;mi&gt;T&lt;/mi&gt;&lt;mi&gt;r&lt;/mi&gt;&lt;mi&gt;a&lt;/mi&gt;&lt;mi&gt;n&lt;/mi&gt;&lt;mi&gt;s&lt;/mi&gt;&lt;mi&gt;X&lt;/mi&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mi&gt;n&lt;/mi&gt;&lt;mi&gt;e&lt;/mi&gt;&lt;mi&gt;t&lt;/mi&gt;&lt;/mrow&gt;&lt;/msub&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt; model extracts mul","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"188 ","pages":"Article 109891"},"PeriodicalIF":7.0,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143479966","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":"Detrended fluctuation analysis of day and night breathing parameters from a wearable respiratory holter","authors":"Alessandra Angelucci,&nbsp;Andrea Aliverti","doi":"10.1016/j.compbiomed.2025.109907","DOIUrl":"10.1016/j.compbiomed.2025.109907","url":null,"abstract":"<div><h3>Background and objective</h3><div>This study focuses on the application of Detrended Fluctuation Analysis (DFA) to understand the variability and correlation properties of respiratory parameters time series obtained by means of a wearable.</div></div><div><h3>Methods</h3><div>Data from 18 healthy volunteers collected using the Airgo™ band, which provides signals proportional to thoracic circumference at a sampling frequency of 10 Hz. The primary aim was to provide preliminary normative data for DFA scaling factors.</div></div><div><h3>Results</h3><div>DFA was applied to 6-h recordings, revealing significant differences (p &lt; 0.001) in scaling factors (α values) for tidal volume (night: 0.97 [0.09], day: 0.88 [0.04]), minute ventilation (night: 1.02 [0.10], day: 0.91 [0.07), mean inspiratory flow (night: 0.98 [0.06], day: 0.88 [0.06]), mean expiratory flow (night: 0.89 [0.08], day: 0.81 [0.06]), and duty cycle (night: 0.64 [0.04], day: 0.59 [0.03]). Quadratic detrending highlighted additional differences not captured with linear detrending, particularly in inspiratory and expiratory time. These findings suggest distinct regulatory patterns during sleep.</div></div><div><h3>Conclusions</h3><div>DFA analysis of respiratory parameters obtained from wearable devices reveals distinct regulatory patterns between day and night conditions, particularly in parameters related to tidal volume and ventilation. These findings demonstrate the potential of DFA to uncover physiological differences in respiratory control mechanisms, especially during sleep, despite technical limitations such as the strong dependency of DFA scaling factors on sampling frequency, duration, and detrending order. Future research should address the limitations of sample size and expand normative datasets to include individuals with respiratory conditions, to translate this methodology into specific clinical applications.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"188 ","pages":"Article 109907"},"PeriodicalIF":7.0,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143480070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced single-cell RNA-seq embedding through gene expression and data-driven gene-gene interaction integration","authors":"Hojjat Torabi Goudarzi ,&nbsp;Maziyar Baran Pouyan","doi":"10.1016/j.compbiomed.2025.109880","DOIUrl":"10.1016/j.compbiomed.2025.109880","url":null,"abstract":"<div><div>Single-cell RNA sequencing (scRNA-seq) provides unprecedented insights into cellular heterogeneity, enabling detailed analysis of complex biological systems at single-cell resolution. However, the high dimensionality and technical noise inherent in scRNA-seq data pose significant analytical challenges. While current embedding methods focus primarily on gene expression levels, they often overlook crucial gene-gene interactions that govern cellular identity and function. To address this limitation, we present a novel embedding approach that integrates both gene expression profiles and data-driven gene-gene interactions. Our method first constructs a Cell-Leaf Graph (CLG) using random forest models to capture regulatory relationships between genes, while simultaneously building a K-Nearest Neighbor Graph (KNNG) to represent expression similarities between cells. These graphs are then combined into an Enriched Cell-Leaf Graph (ECLG), which serves as input for a graph neural network to compute cell embeddings. By incorporating both expression levels and gene-gene interactions, our approach provides a more comprehensive representation of cellular states. Extensive evaluation across multiple datasets demonstrates that our method enhances the detection of rare cell populations and improves downstream analyses such as visualization, clustering, and trajectory inference. This integrated approach represents a significant advance in single-cell data analysis, offering a more complete framework for understanding cellular diversity and dynamics.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"188 ","pages":"Article 109880"},"PeriodicalIF":7.0,"publicationDate":"2025-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143479962","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":"Self-supervised learning for graph-structured data in healthcare applications: A comprehensive review","authors":"Safa Ben Atitallah ,&nbsp;Chaima Ben Rabah ,&nbsp;Maha Driss ,&nbsp;Wadii Boulila ,&nbsp;Anis Koubaa","doi":"10.1016/j.compbiomed.2025.109874","DOIUrl":"10.1016/j.compbiomed.2025.109874","url":null,"abstract":"<div><div>The increasing complexity and interconnectedness of healthcare data present numerous opportunities to improve prediction, diagnosis, and treatment. Graph-structured data, which represents entities and their relationships, is well-suited for modeling these complex connections. However, effectively utilizing this data often requires strong and efficient learning algorithms, especially when dealing with limited labeled data. Self-supervised learning (SSL) has emerged as a powerful paradigm for leveraging unlabeled data to learn effective representations. This paper presents a comprehensive review of SSL approaches specifically designed for graph-structured data in healthcare applications. We explore the challenges and opportunities associated with healthcare data and assess the effectiveness of SSL techniques in real-world healthcare applications. Our discussion encompasses various healthcare settings, such as disease prediction, medical image analysis, and drug discovery. We critically evaluate the performance of different SSL methods across these tasks, highlighting their strengths, limitations, and potential future research directions. To the best of our knowledge, this is the first comprehensive review of SSL applied to graph data in healthcare, providing valuable guidance for researchers and practitioners looking to leverage these techniques to enhance outcomes and drive progress in the field.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"188 ","pages":"Article 109874"},"PeriodicalIF":7.0,"publicationDate":"2025-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143474791","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}