Computer methods and programs in biomedicine最新文献

{"title":"Differentiating characteristics of EMG signals in pediatric muscle tone disorders in the aspect of evaluating postural control","authors":"Aleksandra Tuszy ,&nbsp;Patrycja Romaniszyn-Kania ,&nbsp;Damian Kania ,&nbsp;Andrzej Myśliwiec ,&nbsp;Andrzej Mitas","doi":"10.1016/j.cmpb.2025.108910","DOIUrl":"10.1016/j.cmpb.2025.108910","url":null,"abstract":"<div><h3>Background and Objective:</h3><div>Abnormalities in muscle tone, such as postural hypotonia, can significantly affect motor development and postural control in children, often presenting with unclear origins and subtle clinical manifestations. These disturbances may also be associated with broader musculoskeletal dysfunctions. The presented research aims to examine whether electromyographic signal analysis can support the objective evaluation of muscle tone abnormalities in children.</div></div><div><h3>Methods:</h3><div>Electromyography (EMG) signals were recorded from the sternocleidomastoid and rectus abdominis muscles during the Neck Flexor Endurance Test in 31 children. Time-domain and frequency-domain characteristics were analyzed using statistical methods to differentiate groups classified by physiotherapy experts. Machine learning methods were used to objectively verify the usefulness of the collected data in classification tasks. Statistical analysis included group comparison using Student’s t-test or non-parametric Mann–Whitney U test, where applicable.</div></div><div><h3>Results:</h3><div>Compensatory mechanisms were observed in children with reduced muscle tone, with increased activation of the rectus abdominis muscles. EMG analysis revealed that the rectus abdominis muscles exhibited 25 statistically important features. Feature selection methods like RefielF presented the most differentiating set from sternocleidomastoid muscles (20 features). The Support Vector Machine showed the best overall performance (78.8%) with mean value data set.</div></div><div><h3>Conclusions:</h3><div>The EMG signal analysis revealed significant differences between children with reduced muscle tone and those with normal tone, emphasizing its clinical relevance for pediatric rehabilitation. The promising performance of the tested models suggests that this line of research may be warranted. These findings lay the groundwork for future work and underscore the need for further research on a larger sample to confirm and refine these observations.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"270 ","pages":"Article 108910"},"PeriodicalIF":4.9,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144523406","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":"DC-MSSFF Net: Dule-channel multi-scale spatial-spectral feature fusion network for cholangiocarcinoma pathology high-resolution hyperspectral image segmentation","authors":"Meiyan Liang ,&nbsp;Zelin Xi ,&nbsp;Bo Li ,&nbsp;Lin Wang","doi":"10.1016/j.cmpb.2025.108905","DOIUrl":"10.1016/j.cmpb.2025.108905","url":null,"abstract":"<div><h3>Background and Objective:</h3><div>High-precision segmentation of pathological images is a challenging task in the field of medical image processing. Hyperspectral microscopic imaging offers a distinct advantage in histopathological image segmentation due to its abundance of spectral and spatial data.</div></div><div><h3>Methods:</h3><div>Here, a Dule-Channel Multi-Scale Spatial-Spectral Feature Fusion Network (DC-MSSFF Net) is proposed for semantic segmentation of cholangiocarcinoma hyperspectral images (HSI). The DC-MSSFF Net is composed of two parallel channels, graph-within-graph (GwG) and multi-scale CNN. The GwG can greatly reduce the computational burden while establishing the spatial context relationship of the HSI image. The multi-scale CNN channel is able to fine-tune the segmented edges of the HSI images at the pixel-level based on hyperspectral information in the depth dimension. Afterwards, the segmentation results are achieved by fusing the features from the two channels. Furthermore, an ensemble-based framework is applied to further improve the performance of the model.</div></div><div><h3>Results:</h3><div>The image segmentation evaluation indexes such as dice similarity coefficient (Dice) of the Cholangiocarcinoma HSI data can reach 70.47, which is much higher than the SOTA method and RGB-based image segmentation methods.</div></div><div><h3>Conclusion:</h3><div>The superior performance of the DC-MSSFF network pioneers the inductive learning task of deep frameworks for semantic segmentation of high-resolution hyperspectral image (HR-HSI).</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"269 ","pages":"Article 108905"},"PeriodicalIF":4.9,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144364716","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":"SmartAlert: Machine learning-based patient-ventilator asynchrony detection system in intensive care units","authors":"Jaroslav Pažout ,&nbsp;Milan Němý ,&nbsp;Jakub Mikeš ,&nbsp;Jan Jirman ,&nbsp;Jan Kubr ,&nbsp;Eliška Niebauerová ,&nbsp;Miroslav Macík ,&nbsp;Michal Pech ,&nbsp;Michal Štajnrt ,&nbsp;Jakub Vaněk ,&nbsp;Petr Waldauf ,&nbsp;Václav Zvoníček ,&nbsp;Lenka Vysloužilová ,&nbsp;Robert Babuška ,&nbsp;František Duška ,&nbsp;VentConnect Study group","doi":"10.1016/j.cmpb.2025.108927","DOIUrl":"10.1016/j.cmpb.2025.108927","url":null,"abstract":"<div><h3>Background and Objective</h3><div>Patient-ventilator asynchronies (PVA) are associated with ventilator-induced lung injury and increased mortality. Current detection methods rely on static thresholds, extensive preprocessing, or proprietary ventilator data. This study aimed to develop and validate a fully online, real-time system that detects and classifies PVAs directly from ventilator screen data while alerting clinicians based on severity.</div></div><div><h3>Methods</h3><div>The SmartAlert system was developed using ventilator screen recordings from ICU patients. It extracts pressure and flow waveforms from video recordings, converts them into time-series data, and employs deep neural networks to classify asynchronies and assign alarm levels from no urgency to most urgent. A dataset of 381,280 double-breath units was independently annotated by two expert intensivists. Two deep learning models were trained: one for alarm prediction and another for asynchrony classification (ineffective triggering, double cycling, high inspiratory effort, no asynchrony). Performance was evaluated using accuracy, sensitivity, specificity, and AUC-ROC, compared to expert consensus.</div></div><div><h3>Results</h3><div>SmartAlert demonstrated strong performance for alarm level prediction (overall accuracy: 83.8 %, weighted AUC-ROC: 0.943 [95 % CI: 0.941–0.945]) and PVA classification (weighted accuracy: 89.3 %, weighted AUC-ROC: 0.951 [95 % CI: 0.950–0.953]). It showed high specificity for urgent alarms (99.9 % for level 3) and PVA types (98.5 % for ineffective triggering, 96.9 % for double cycling, 94.8 % for high inspiratory effort).</div></div><div><h3>Conclusions</h3><div>We developed and internally validated SmartAlert, an automated system that detects PVAs, classifies severity, and alerts clinicians in real time. Its potential to reduce alarm fatigue, optimize ventilator settings, and improve patient outcomes remains to be tested in clinical trials.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"269 ","pages":"Article 108927"},"PeriodicalIF":4.9,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144501925","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":"Stabilizing machine learning for reproducible and explainable results: A novel validation approach to subject-specific insights","authors":"Gideon Vos ,&nbsp;Liza van Eijk ,&nbsp;Zoltan Sarnyai ,&nbsp;Mostafa Rahimi Azghadi","doi":"10.1016/j.cmpb.2025.108899","DOIUrl":"10.1016/j.cmpb.2025.108899","url":null,"abstract":"&lt;div&gt;&lt;h3&gt;Introduction:&lt;/h3&gt;&lt;div&gt;Machine Learning (ML) is transforming medical research by enhancing diagnostic accuracy, predicting disease progression, and personalizing treatments. While general models trained on large datasets identify broad patterns across populations, the diversity of human biology, shaped by genetics, environment, and lifestyle, often limits their effectiveness. This has driven a shift towards subject-specific models that incorporate individual biological and clinical data for more precise predictions and personalized care. However, developing these models presents significant practical and financial challenges. Additionally, ML models initialized through stochastic processes with random seeds can suffer from reproducibility issues when those seeds are changed, leading to variations in predictive performance and feature importance. To address this, this study introduces a novel validation approach to enhance model interpretability, stabilizing predictive performance and feature importance at both the group and subject-specific levels.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Methods:&lt;/h3&gt;&lt;div&gt;We conducted initial experiments using a single Random Forest (RF) model initialized with a random seed for key stochastic processes, on nine datasets that varied in domain problems, sample size, and demographics. Different validation techniques were applied to assess model accuracy and reproducibility while evaluating feature importance consistency. Next, the experiment was repeated for each dataset for up to 400 trials per subject, randomly seeding the machine learning algorithm between each trial. This introduced variability in the initialization of model parameters, thus providing a more comprehensive evaluation of the machine learning model’s features and performance consistency. The repeated trials generated up to 400 feature sets per subject. By aggregating feature importance rankings across trials, our method identified the most consistently important features, reducing the impact of noise and random variation in feature selection. The top subject-specific feature importance set across all trials was then identified. Finally, using all subject-specific feature sets, the top group-specific feature importance set was also created. This process resulted in stable, reproducible feature rankings, enhancing both subject-level and group-level model explainability.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Results:&lt;/h3&gt;&lt;div&gt;We found that machine learning models with stochastic initialization were particularly susceptible to variations in reproducibility, predictive accuracy, and feature importance due to random seed selection and validation techniques during training. Changes in random seeds altered weight initialization, optimization paths, and feature rankings, leading to fluctuations in test accuracy and interpretability. These findings align with prior research on the sensitivity of stochastic models to initialization randomness. This study builds on that understanding by i","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"269 ","pages":"Article 108899"},"PeriodicalIF":4.9,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144470432","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":"Optimizing beat-wise input for arrhythmia detection using 1-D convolutional neural networks: A real-world ECG study","authors":"Sunghan Lee ,&nbsp;Guangyao Zheng ,&nbsp;Jeonghwan Koh ,&nbsp;Haoran Li ,&nbsp;Zicheng Xu ,&nbsp;Sung Pil Cho ,&nbsp;Sung Il Im ,&nbsp;Vladimir Braverman ,&nbsp;In cheol Jeong","doi":"10.1016/j.cmpb.2025.108898","DOIUrl":"10.1016/j.cmpb.2025.108898","url":null,"abstract":"<div><h3>Backgrounds and objectives:</h3><div>Cardiac arrhythmias, characterized by irregular heartbeats, are difficult to diagnose in real-world scenarios. Machine learning has advanced arrhythmia detection; however, the optimal number of heartbeats for precise classification remains understudied. This study addresses this using machine learning while assessing the performance of arrhythmia detection across inter-patient and intra-patient conditions. Furthermore, the performance–resource trade-offs are evaluated for practical deployment in mobile health (mHealth) applications.</div></div><div><h3>Methods:</h3><div>Beat-wise segmentation and resampling techniques were utilized for preprocessing electrocardiography (ECG) signals to ensure consistent input lengths. A 1-D convolutional neural network was used to classify the eight multi-labeled arrhythmias. The dataset comprised real-world ECG recordings from the HiCardi wireless device alongside data from the MIT-BIH Arrhythmia database. Model performance was assessed through fivefold cross-validation under both inter-patient and intra-patient conditions.</div></div><div><h3>Results:</h3><div>The proposed model demonstrated peak accuracy at four beats under inter-patient conditions, with minimal improvements beyond this point. This configuration achieved a balance between performance (94.82% accuracy) and resource consumption (training time: 72.27 s per epoch; prediction time: 155 <span><math><mi>μ</mi></math></span>s per segment). Real-world simulations validated the feasibility of real-time arrhythmia detection for approximately 5000 patients.</div></div><div><h3>Conclusion:</h3><div>Utilizing four heartbeats as the input size for arrhythmia classification results in a trade-off between accuracy and computational efficiency. This discovery has significant implications for real-time wearable ECG devices, where both performance and resource constraints are crucial considerations. This insight is expected to serve as a valuable reference for enhancing the design and implementation of arrhythmia detection systems for scalable and efficient mHealth applications.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"269 ","pages":"Article 108898"},"PeriodicalIF":4.9,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144338982","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":"Designing a Computer-Aided Detection system for Barrett ’s neoplasia: Insights in architectural choices, training strategies and inference approaches","authors":"Carolus H.J. Kusters ,&nbsp;Tim G.W. Boers ,&nbsp;Tim J.M. Jaspers ,&nbsp;Martijn R. Jong ,&nbsp;Rixta A.H. van Eijck van Heslinga ,&nbsp;Jelmer B. Jukema ,&nbsp;Kiki N. Fockens ,&nbsp;Albert J. de Groof ,&nbsp;Jacques J. Bergman ,&nbsp;Fons van der Sommen ,&nbsp;Peter H.N. De With","doi":"10.1016/j.cmpb.2025.108891","DOIUrl":"10.1016/j.cmpb.2025.108891","url":null,"abstract":"&lt;div&gt;&lt;h3&gt;Background and Objective:&lt;/h3&gt;&lt;div&gt;Detecting early neoplasia in Barrett’s Esophagus (BE) presents significant challenges due to the subtle endoscopic appearance of lesions. Computer-Aided Detection (CADe) systems have the potential to assist endoscopists by enhancing the identification and localization of these early-stage lesions. This study aims to provide comprehensive insights into the structured design and development of effective CADe systems for BE neoplasia detection, addressing unique challenges and complexities of endoscopic imaging and the nature of BE neoplasia.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Methods:&lt;/h3&gt;&lt;div&gt;We conduct an extensive evaluation of architectural choices, training strategies, and inference approaches to optimize CADe systems for BE neoplasia detection. This evaluation includes 10 backbone architectures and 4 semantic segmentation decoders. Training strategies assessed are domain-specific pre-training with a self-supervised learning objective, data augmentation techniques, incorporation of additional video frames and utilization of variants for multi-expert segmentation ground-truth. Evaluation of inference approaches includes various model output fusion techniques and TensorRT conversion. The optimized model is benchmarked against 6 state-of-the-art CADe systems for BE neoplasia detection across 9 diverse test sets.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Results:&lt;/h3&gt;&lt;div&gt;The experimental results demonstrate the impact of incorporating structured design considerations, leading to measurable and incremental performance gains of up to 7.8% on dedicated validation sets. The contributions particularly stand out for the domain-specific pre-training and the use of a hybrid CNN-Transformer architecture, which benefits robustness and overall performance. The model optimized through these design choices achieves statistically significant improvements over existing CADe systems, with &lt;span&gt;&lt;math&gt;&lt;mi&gt;p&lt;/mi&gt;&lt;/math&gt;&lt;/span&gt;-values in the range &lt;span&gt;&lt;math&gt;&lt;mrow&gt;&lt;mi&gt;p&lt;/mi&gt;&lt;mo&gt;∈&lt;/mo&gt;&lt;mrow&gt;&lt;mo&gt;[&lt;/mo&gt;&lt;mn&gt;0&lt;/mn&gt;&lt;mo&gt;.&lt;/mo&gt;&lt;mn&gt;0019&lt;/mn&gt;&lt;mo&gt;,&lt;/mo&gt;&lt;mn&gt;0&lt;/mn&gt;&lt;mo&gt;.&lt;/mo&gt;&lt;mn&gt;031&lt;/mn&gt;&lt;mo&gt;]&lt;/mo&gt;&lt;/mrow&gt;&lt;/mrow&gt;&lt;/math&gt;&lt;/span&gt;. It outperforms state-of-the-art models in classification and localization, with improvements of up to 12.8% over the second-best performing model. These gains demonstrate enhanced peak performance, generalization capabilities, and robustness across diverse test sets representative of real-world clinical challenges.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Conclusion:&lt;/h3&gt;&lt;div&gt;This study provides critical insights into the structured development of effective CADe systems for Barrett’s neoplasia detection. By addressing the specific challenges associated with endoscopic imaging and Barrett’s neoplasia, the study demonstrates that careful consideration of architectural choices, training strategies, and inference approaches results in significantly improved CADe performance. These findings underscore the importance of tailored design and optimization in developing robust an","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"269 ","pages":"Article 108891"},"PeriodicalIF":4.9,"publicationDate":"2025-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144364677","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":"Machine learning strategies for multi-label pre-diagnosis of diseases with superficial data","authors":"Dengqun Gou ,&nbsp;Xu Luo ,&nbsp;Zhichen Liu","doi":"10.1016/j.cmpb.2025.108911","DOIUrl":"10.1016/j.cmpb.2025.108911","url":null,"abstract":"<div><h3>Background and objective</h3><div>General practice (GP) pre-diagnosis, a key task in disease triage, directs patients to suitable departments despite limited data and multi-label classification challenges. To address this issue, a framework with dimensionality reduction machine learning strategies was provided.</div></div><div><h3>Methods</h3><div>Disease information was organized into hierarchical tiers, focusing primarily on overarching disease classifications (I-level) and their subcategories (II-level). Two machine learning strategies were introduced and embedded into a framework. One was the classifier chain strategy, and the other one was ensemble learning-DNN (Deep Neural Networks) strategy. In classifier chains, the base candidate algorithms included XGBoost, RF (Random Forest), LR (Logistic Regression), and SVM (Support Vector Machine). In GP pre-diagnosis, the I-level and II-level disease information was progressively inferred. The efficacy of the methodologies was demonstrated through 3124 retrospective electronic medical records of patients complaining of abdominal pain. The performance metrics included AUPRC, AUROC, F1, accuracy, sensitivity, specificity, and hamming loss. The performance of different machine learning approaches was compared using the Friedman test, followed by the Nemenyi post-hoc test.</div></div><div><h3>Results</h3><div>The statistical results indicated that the Classifier chain-RF approach was optimal. For overarching disease categorizations, performance was excellent with nearly all metrics exceeding 0.90. For disease subcategories, performance slightly declined but remained highly effective, with most metrics surpassing 0.80.</div></div><div><h3>Conclusions</h3><div>The proposed framework exhibited its efficacy by performing well across various metrics and successfully accomplishing the established objectives, contributing insights to computer-aided diagnosis in the specific area of GP pre-diagnosis. Classifier chain-RF is recommended as an embedding approach.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"269 ","pages":"Article 108911"},"PeriodicalIF":4.9,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144322393","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":"Unraveling the tumor-microenvironment through a radiogenomic-based multiomic approach to predict outcomes of immunotherapy in non-small cell lung cancer","authors":"Dong Young Jeong ,&nbsp;Cheol Yong Joe ,&nbsp;Sang Min Lee ,&nbsp;Sehhoon Park ,&nbsp;Seung Hwan Moon ,&nbsp;Joon Young Choi ,&nbsp;Jonghoon Kim ,&nbsp;Se-Hoon Lee ,&nbsp;Ho Yun Lee","doi":"10.1016/j.cmpb.2025.108915","DOIUrl":"10.1016/j.cmpb.2025.108915","url":null,"abstract":"<div><h3>Background</h3><div>The tumor microenvironment (TME) plays a critical role in influencing immune checkpoint inhibitor (ICI) therapy outcomes in advanced non-small cell lung cancer (NSCLC). This study aimed to develop a radiomics model reflecting an ICI-favorable TME based on whole transcriptome sequencing (WTS).</div></div><div><h3>Methods</h3><div>This multi-center retrospective cohort study included training (n = 120), internal validation (n = 319), and external validation (n = 150) cohorts of advanced NSCLC patients who received ICI as first- or second-line therapy. The radiomics model (rTME) was developed based on the TME score, which reflected ICI-favorable immune cell compositions. The model’s performance was assessed using the C-index, and survival outcomes were also evaluated.</div></div><div><h3>Results</h3><div>In the training cohort, high rTME scores were associated with significantly prolonged progression-free survival (PFS) (median 4.1 vs. 2.9 months, p = 0.024) and overall survival (OS) (median 15.0 vs. 8.4 months, p = 0.030). Similar trends were observed in the internal validation cohort for PFS (median 3.3 vs. 2.1 months, p = 0.004) and OS (median 13.9 vs. 7.3 months, p = 0.004), as well as in the external validation cohort for OS (median 15.5 vs. 7.3 months, p = 0.008). Integrating clinical variables improved predictive accuracy in both the training and internal validation cohorts.</div></div><div><h3>Conclusion</h3><div>Our radiomics model, reflecting the ICI-favorable immune cell expression in the TME, showed a positive association with ICI outcomes in NSCLC patients. Integrating radiomics and clinical variables enhances prognostic accuracy, demonstrating the model’s potential utility in guiding ICI therapy decisions.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"269 ","pages":"Article 108915"},"PeriodicalIF":4.9,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144522181","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":"Weakly-supervised segmentation using sparse single point annotations for lumen and wall of carotid arteries in 3D MRI","authors":"Jonghun Kim ,&nbsp;Inye Na ,&nbsp;Junmo Kwon ,&nbsp;Woo-Keun Seo ,&nbsp;Hyunjin Park","doi":"10.1016/j.cmpb.2025.108881","DOIUrl":"10.1016/j.cmpb.2025.108881","url":null,"abstract":"<div><h3>Background and objective:</h3><div>Segmentation of the carotid artery is a crucial step in planning therapy for atherosclerosis. Manual annotation is a time-consuming and labor-intensive process, and there is a need to reduce this effort.</div></div><div><h3>Methods:</h3><div>We propose a weakly supervised segmentation method using only a few annotated axial slices, each with a single-point annotation from 3D magnetic resonance imaging for the lumen and wall of the carotid artery. The proposed method contains three loss functions designed to (1) locate the center point of the vessel, (2) constrain the range of the vessel radius using prior information implemented with spatial maps, and (3) encourage similar segmentation results in adjacent slices. Both the lumen (inner structure) and wall (outer structure) can be segmented by adjusting the range of plausible radii.</div></div><div><h3>Results:</h3><div>Experimental evaluations on the COSMOS2022 dataset show that our method achieved similar performance results (<span><math><msub><mrow><mtext>DSC</mtext></mrow><mrow><mi>l</mi><mi>u</mi></mrow></msub></math></span> 0.821 lumen, <span><math><msub><mrow><mtext>DSC</mtext></mrow><mrow><mi>w</mi><mi>a</mi></mrow></msub></math></span> 0.841 wall) to those of fully supervised methods with dense annotations (<span><math><msub><mrow><mtext>DSC</mtext></mrow><mrow><mi>l</mi><mi>u</mi></mrow></msub></math></span> 0.814-0.857, <span><math><msub><mrow><mtext>DSC</mtext></mrow><mrow><mi>w</mi><mi>a</mi></mrow></msub></math></span> 0.832-0.875). Similar trends were observed on an independent Harvard dataset.</div></div><div><h3>Conclusion:</h3><div>Our proposed method demonstrated effective segmentation of crucial arteries, internal carotid artery, external carotid artery, and common carotid artery in atherosclerosis. We anticipate that this efficient approach utilizing single-point annotation will contribute to the effective management of carotid atherosclerosis. Our code is available at <span><span>https://github.com/jongdory/CASCA</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"269 ","pages":"Article 108881"},"PeriodicalIF":4.9,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144307337","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 quantitative analysis of peri-articular bone microarchitecture in HR-pQCT knee images","authors":"Nathan J. Neeteson ,&nbsp;Sasha M. Hasick ,&nbsp;Roberto Souza ,&nbsp;Steven K. Boyd","doi":"10.1016/j.cmpb.2025.108882","DOIUrl":"10.1016/j.cmpb.2025.108882","url":null,"abstract":"&lt;div&gt;&lt;div&gt;Applying HR-pQCT to image the knee necessitates the development and validation of novel image analysis workflows. Here, we present and validate the first automated workflow for &lt;em&gt;in vivo&lt;/em&gt; quantitative assessment of peri-articular bone density and microarchitecture in the knee. Segmentation models were first trained with radius and tibia images (N=2,598) then fine-tuned with knee images (N=131). Atlas-based registration was used to create medial and lateral contact surface masks, which were combined with bone segmentations to generate peri-articular regions of interest masks. The accuracy and precision of the workflow was assessed with an external validation dataset (N=128) and a triple-repeat measures dataset (N=29), respectively. Predicted and reference morphological parameters had linear coefficients of determination between 0.86 and 0.99, with moderate bias present in predictions of subchondral bone plate density and thickness. The average short-term precision RMS%CV estimates across all compartments and all morphological parameters ranged from 1.0 % to 2.9 %.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Background and Objective:&lt;/h3&gt;&lt;div&gt;There is growing interest in applying HR-pQCT to image the knee, particularly in the study of osteoarthritis. This necessitates the development and validation of novel image analysis workflows tailored to knee HR-pQCT images. In this work, we present and validate the first fully automated workflow for &lt;em&gt;in vivo&lt;/em&gt; quantitative assessment of peri-articular bone density and microarchitecture in the human knee.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Methods:&lt;/h3&gt;&lt;div&gt;Bone segmentation models were trained by transfer learning with a large dataset of radius and tibia images (N=2,598) and fine-tuned on a knee image dataset (N=131). Tibia and femur atlases were created and atlas-based registration was used to identify medial and lateral contact surfaces. Morphological operations combined bone segmentations and atlas-generated contact surface masks to generate peri-articular regions of interest masks, in which standard morphological analysis was applied. The accuracy and precision of estimated morphological parameters was assessed with an external validation dataset containing femurs and tibiae (N=128) and a triple-repeat measures dataset containing only tibiae (N=29), respectively.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Results:&lt;/h3&gt;&lt;div&gt;On the external validation dataset, predicted and reference morphological parameters showed excellent correspondence (0.86 &lt;span&gt;&lt;math&gt;&lt;mo&gt;≤&lt;/mo&gt;&lt;/math&gt;&lt;/span&gt; R&lt;span&gt;&lt;math&gt;&lt;msup&gt;&lt;mrow&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mn&gt;2&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;/math&gt;&lt;/span&gt; &lt;span&gt;&lt;math&gt;&lt;mo&gt;≤&lt;/mo&gt;&lt;/math&gt;&lt;/span&gt; 0.99), with moderate bias present in predictions of subchondral bone plate density (−80 mg HA/cm&lt;span&gt;&lt;math&gt;&lt;msup&gt;&lt;mrow&gt;&lt;/mrow&gt;&lt;mrow&gt;&lt;mn&gt;3&lt;/mn&gt;&lt;/mrow&gt;&lt;/msup&gt;&lt;/math&gt;&lt;/span&gt;) and thickness (＋0.15 mm). With intra-participant rigid registration, the average short-term precision RMS%CV estimates across all compartments were 2.2 % and 2.8 % for subchondral bone ","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"269 ","pages":"Article 108882"},"PeriodicalIF":4.9,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144297376","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}