Computers in biology and medicine最新文献

{"title":"Consensus-guided evaluation of self-supervised learning in echocardiographic segmentation","authors":"Preshen Naidoo ,&nbsp;Patricia Fernandes ,&nbsp;Nasim Dadashi Serej ,&nbsp;Charlotte H. Manisty ,&nbsp;Matthew J. Shun-Shin ,&nbsp;Darrel P. Francis ,&nbsp;Massoud Zolgharni","doi":"10.1016/j.compbiomed.2025.111148","DOIUrl":"10.1016/j.compbiomed.2025.111148","url":null,"abstract":"<div><h3>Background:</h3><div>Left ventricle segmentation is a fundamental task in echocardiography, essential for assessing cardiac function. However, deep learning models for segmentation rely on large labelled datasets, which are expensive and time-consuming to annotate. Self-supervised learning has emerged as a promising approach to leverage unlabelled data, but its effectiveness for left ventricle segmentation remains underexplored.</div></div><div><h3>Methods:</h3><div>This study investigates self-supervised learning for echocardiographic segmentation, comparing various pretext tasks. The impact of dataset size and distribution on pre-training is examined, revealing that excessive unlabelled data can degrade performance due to redundancy and low variability. A novel multi-expert labelled dataset is introduced to enhance segmentation evaluation, using consensus-based annotations to reduce annotation noise and improve reliability.</div></div><div><h3>Results:</h3><div>Among the self-supervised learning methods evaluated, contrastive learning consistently outperforms other approaches, particularly in low-label settings. The study demonstrates that AI models pre-trained using self-supervised learning and fine-tuned with only 15% of labelled data achieve stronger alignment with multi-expert consensus than any individual expert.</div></div><div><h3>Conclusion:</h3><div>The findings suggest that AI models can generalise well across expert annotations, providing more reliable and reproducible assessments.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"198 ","pages":"Article 111148"},"PeriodicalIF":6.3,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145211916","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":"Building digital twins for personalized cardiovascular medicine: Advances, challenges, and future directions.","authors":"Claudio Chiastra, Selene Pirola, Simone Saitta, Francesco Sturla, John F LaDisa","doi":"10.1016/j.compbiomed.2025.111122","DOIUrl":"https://doi.org/10.1016/j.compbiomed.2025.111122","url":null,"abstract":"","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":" ","pages":"111122"},"PeriodicalIF":6.3,"publicationDate":"2025-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145206004","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":"Non-contrast CT-based pulmonary embolism detection using GAN-generated synthetic contrast enhancement: Development and validation of an AI framework","authors":"Young-Tak Kim ,&nbsp;So Hyeon Bak ,&nbsp;Seon-Sook Han ,&nbsp;Yunsik Son ,&nbsp;Jinkyeong Park","doi":"10.1016/j.compbiomed.2025.111109","DOIUrl":"10.1016/j.compbiomed.2025.111109","url":null,"abstract":"<div><div>Acute pulmonary embolism (PE) is a life-threatening condition often diagnosed using CT pulmonary angiography (CTPA). However, CTPA is contraindicated in patients with contrast allergies or at risk for contrast-induced nephropathy. This study explores an AI-driven approach to generate synthetic contrast-enhanced images from non-contrast CT scans for accurate diagnosis of acute PE without contrast agents. This retrospective study used dual-energy and standard CT datasets from two institutions. The internal dataset included 84 patients: 41 PE-negative cases for generative model training and 43 patients (30 PE-positive) for diagnostic evaluation. An external dataset of 62 patients (26 PE-positive) was used for further validation. We developed a generative adversarial network (GAN) based on U-Net, trained on paired non-contrast and contrast-enhanced images. The model was optimized using contrast-enhanced L1-loss with hyperparameter λ to improve anatomical accuracy. A ConvNeXt-based classifier trained on the RSNA dataset (N = 7,122) generated per-slice PE probabilities, which were aggregated for patient-level prediction via a Random Forest model. Diagnostic performance was assessed using five-fold cross-validation on both internal and external datasets. The GAN achieved optimal image similarity at λ = 0.5, with the lowest mean absolute error (0.0089) and highest MS-SSIM (0.9674). PE classification yielded AUCs of 0.861 and 0.836 in the internal dataset, and 0.787 and 0.680 in the external dataset, using real and synthetic images, respectively. No statistically significant differences were observed. Our findings demonstrate that synthetic contrast CT can serve as a viable alternative for PE diagnosis in patients contraindicated for CTPA, supporting safe and accessible imaging strategies.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"198 ","pages":"Article 111109"},"PeriodicalIF":6.3,"publicationDate":"2025-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145206007","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":"Advancement in hepatocellular carcinoma research: Biomarkers, therapeutics approaches and impact of artificial intelligence","authors":"Devraj Rajak,&nbsp;Priyanshu Nema,&nbsp;Arvindra Sahu,&nbsp;Satyamshyam Vishwakarma,&nbsp;Sushil K. Kashaw","doi":"10.1016/j.compbiomed.2025.111120","DOIUrl":"10.1016/j.compbiomed.2025.111120","url":null,"abstract":"<div><div>Cancer is a leading, highly complex, and deadly disease that has become a major concern in modern medicine. Hepatocellular carcinoma is the most common primary liver cancer and a leading cause of global cancer mortality. Its development is predominantly associated with chronic liver diseases such as hepatitis B and C infections, cirrhosis, alcohol consumption, and non-alcoholic fatty liver disease. Molecular mechanisms underlying HCC involve genetic mutations, epigenetic changes, and disrupted signalling pathways, including Wnt/β-catenin and PI3K/AKT/mTOR. Early diagnosis remains challenging, as most cases are detected at advanced stages, limiting curative treatment options. Diagnostic advancements, including biomarkers like alpha-fetoprotein and cutting-edge imaging techniques such as CT, MRI, and ultrasound-based radiomics, have improved early detection. Treatment strategies depend on the disease stage, ranging from curative options like surgical resection and liver transplantation to palliative therapies, including transarterial chemoembolization, systemic therapies, and immunotherapy. Immune checkpoint inhibitors targeting PD-1/PD-L1 and CTLA-4 have shown promise for advanced HCC. In this review we discuss about emerging technologies, including artificial intelligence and multi-omics platforms for HCC management by enhancing diagnostic accuracy, identifying novel therapeutic targets, and enabling personalized treatments. Despite these advancements, the prognosis for HCC patients remains poor, underscoring the need for continued research into early detection, innovative therapies, and translational applications to effectively address this global health challenge.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"198 ","pages":"Article 111120"},"PeriodicalIF":6.3,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145197827","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":"Physiologically constrained neuromuscular synergy extraction using a deep off-policy dynamic neuro-fuzzy system in wheelchair propulsion","authors":"Mohammad Mahdi Rusta","doi":"10.1016/j.compbiomed.2025.111141","DOIUrl":"10.1016/j.compbiomed.2025.111141","url":null,"abstract":"<div><div>Manual wheelchair propulsion (MWP) is a repetitive activity that risks upper limb injuries, necessitating analysis of intramuscular coordination for effective intervention. The state-of-the-art synergy extraction methods struggle with MWP's nonlinear, dynamic nature and often overlook biomechanical constraints. This study introduces a deep reinforcement learning-based dynamic neuro-fuzzy (DRL-DNF) system that models complex electromyography (EMG) patterns and refines synergy structures in real time. Data from 24 manual wheelchair users included EMG signals from six muscles, joint kinematics, and handrim force. A musculoskeletal model was incorporated to account for joint dynamics and external forces, ensuring physiologically meaningful synergy extraction. Statistical analysis showed that DRL-DNF outperformed conventional methods, achieving a mean variance accounted for (VAF) of 94.12 ± 4.12 % in noise-free and 90.1 ± 4.18 % in noise-presence for three synergies, indicating strong robustness to noise and superior modeling of agonist-antagonist interactions. Significant differences in third synergy activations revealed autoencoder (AE) had higher values than independent component analysis (ICA) and non-negative matrix factorization (NMF), while ICA had lower values than NMF and DRL-DNF. Statistical parametric mapping indicated temporal differences, with ICA underactivating during the push and recovery phases, and AE overactivating in late recovery phase. Synergy coefficients also differed significantly across methods. AE consistently assigned lower weights to biarticular muscles, while NMF emphasized shoulder flexors and DRL-DNF provided a balanced representation. DRL-DNF identified physiologically relevant synergies, enhancing insights into intramuscular coordination strategies and showing potential for synergy-based control in assistive devices and rehabilitation. Future work will aim to enhance computational efficiency and expand dataset diversity for real-time application.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"198 ","pages":"Article 111141"},"PeriodicalIF":6.3,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145198215","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":"Non-invasive tidal volume estimation with wearable sensors using a high-gain observer and deep learning","authors":"Meng Ba,&nbsp;Paolo Pianosi,&nbsp;Rajesh Rajamani","doi":"10.1016/j.compbiomed.2025.111114","DOIUrl":"10.1016/j.compbiomed.2025.111114","url":null,"abstract":"<div><div>Non-invasive tidal volume (TV) estimation can be valuable for respiratory monitoring, particularly for patients needing continuous assessment. Traditional spirometry-based methods are precise but impractical for daily use due to their invasive nature, discomfort and limitations on mobility. This study integrates a nonlinear high-gain observer (HGO) with a convolutional neural network long short-term memory network (CNN-LSTM) to estimate TV using wearable inertial measurement unit (IMU) sensors. The HGO provides reliable thoracoabdominal displacements by mitigating sensor drift and removing gravity components measured by the accelerometer. Combined with raw IMU data, these displacements serve as inputs for a deep learning CNN-LSTM network, which captures spatial and temporal dependencies to improve prediction accuracy. The CNN-LSTM model trained with both data sources demonstrated superior accuracy and also a high degree of robustness to sensor placement variations. Experimental results in an IRB approved study with 6 subjects show that the method achieved an averaged RMS error of 40.38 mL even with repeated taking off and re-wearing of the sensors. These findings underscore the potential of replacing invasive spirometry with convenient wearable sensors when coupled with reliable estimation algorithms.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"198 ","pages":"Article 111114"},"PeriodicalIF":6.3,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145198241","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":"Trust and accuracy in AI: Optometrists favor multimodal AI systems over unimodal for glaucoma diagnosis in collaborative environment","authors":"Faisal Ghaffar ,&nbsp;Yousuf Zia Islam ,&nbsp;Nadine Furtado ,&nbsp;Catherine Burns","doi":"10.1016/j.compbiomed.2025.111132","DOIUrl":"10.1016/j.compbiomed.2025.111132","url":null,"abstract":"<div><h3>Background:</h3><div>User trust and decision accuracy are crucial for the successful collaboration of humans and Artificial Intelligence (AI) systems, especially in clinical settings such as glaucoma diagnosis. Both trust and accuracy are influenced by the data modality used by AI systems, which directly impacts the effectiveness of human-AI collaboration.</div></div><div><h3>Objective:</h3><div>The objective of this study is to discover the modality of an AI system that aligns best with an optometrist’s mental model. This was achieved by comparing trust levels between unimodal and multimodal AI systems used for glaucoma diagnosis. Additionally, we explore the impact of modality on various targets of user trust and user performance.</div></div><div><h3>Methods:</h3><div>We conducted a within-subject study with 20 optometrists, who were presented with both unimodal and multimodal AI mock-up systems specifically designed for glaucoma diagnosis. Trust was measured across five key targets using a 5 point Likert scale questionnaires. Statistical analysis was performed to assess trust differences between the unimodal and multimodal AI systems. Optometrist performance was evaluated based on the alignment of their decisions with those of the unimodal and multimodal AI systems.</div></div><div><h3>Results:</h3><div>The results showed that the multimodal system had a higher average trust rating of 3.98 on a Likert scale, indicating greater trust compared to the unimodal system, which had an average trust rating of 3.00. This difference was statistically significant (<em>p</em><span><math><mo>&lt;</mo></math></span>.001), with further analysis revealing significant variation across all trust targets (<em>p</em><span><math><mo>&lt;</mo></math></span>.001). Additionally, optometrists demonstrated higher F1 scores with the multimodal system (43.1%) compared to the unimodal system (37.3%), while accuracy remained comparable between the two systems (63.0% for multimodal and 63.3% for unimodal). However, major differences across these metrics were observed in relation to optometrist’s expertise.</div></div><div><h3>Conclusions:</h3><div>Multimodal AI systems, which use the same data modality as optometrists and align more closely with their mental models and decision-making processes, result in higher user trust and improved diagnostic performance. Therefore, for effective human-AI collaboration in healthcare, specifically for glaucoma diagnosis, AI systems should be designed to utilize the same data sources as the human counterparts, ensuring consistency and improving both trust and decision accuracy.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"198 ","pages":"Article 111132"},"PeriodicalIF":6.3,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145198341","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":"Towards population scale testis volume segmentation in DIXON MRI","authors":"Jan Ernsting ,&nbsp;Philipp Nikolas Beeken ,&nbsp;Lynn Ogoniak ,&nbsp;Jacqueline Kockwelp ,&nbsp;Wolfgang Roll ,&nbsp;Tim Hahn ,&nbsp;Alexander Siegfried Busch ,&nbsp;Benjamin Risse","doi":"10.1016/j.compbiomed.2025.111139","DOIUrl":"10.1016/j.compbiomed.2025.111139","url":null,"abstract":"<div><div>Testis size is known to be one of the main predictors of male fertility, usually assessed in clinical workup via palpation or imaging. Despite its potential, population-level evaluation of testicular volume using imaging remains underexplored. Previous studies, limited by small and biased datasets, have demonstrated the feasibility of machine learning for testis volume segmentation. This paper presents an evaluation of segmentation methods for testicular volume using Magnetic Resonance Imaging data from the UKBiobank. The best model achieves a median dice score of 0.89, compared to median dice score of 0.85 for human interrater reliability on the same dataset, enabling large-scale annotation on a population scale for the first time. Our overall aim is to provide a trained model, comparative baseline methods, and annotated training data to enhance accessibility and reproducibility in testis MRI segmentation research.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"198 ","pages":"Article 111139"},"PeriodicalIF":6.3,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145198179","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":"Comment on ''Enhancing clinical decision support with physiological waveforms - A multimodal benchmark in emergency care\" by Lopez Alcaraz et al.","authors":"Kaiyuan Liu","doi":"10.1016/j.compbiomed.2025.111140","DOIUrl":"10.1016/j.compbiomed.2025.111140","url":null,"abstract":"","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"198 ","pages":"Article 111140"},"PeriodicalIF":6.3,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145198227","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":"Design, synthesis, and pharmacological evaluation of hydrazide–imine diclofenac derivatives with dual anti-inflammatory and anticancer potential","authors":"Lalnun Hruaitluangi ,&nbsp;Lalduhawma Chhakchhuak ,&nbsp;Caroline Malsawmtluangi ,&nbsp;R. Vanlalruata ,&nbsp;Ajmal Koya Pulikkal ,&nbsp;Lalzikpuii Sailo","doi":"10.1016/j.compbiomed.2025.111123","DOIUrl":"10.1016/j.compbiomed.2025.111123","url":null,"abstract":"<div><div>Diclofenac, a widely used nonsteroidal anti-inflammatory drug (NSAID), is effective in treating pain and inflammation and has also shown potential as an anticancer agent, primarily through cyclooxygenase-2 (COX-2) inhibition. However, its therapeutic application is often limited by adverse effects, necessitating the development of new derivatives with improved pharmacological profiles. In this study, two novel diclofenac hydrazide–imine derivatives, DDCH (cyclohexanone-based) and DDAC (acetylacetone-based), were synthesized, structurally characterized, and evaluated through combined computational and experimental approaches. Density functional theory (DFT) calculations provided transition-state and energy profile analyses, while molecular docking and molecular dynamics (MD) simulations established stable interactions of both derivatives with COX-2 and heat shock protein 90 (HSP90), a key oncogenic chaperone. DDAC exhibited particularly strong binding to HSP90, suggesting enhanced anticancer potential compared with diclofenac. <em>In silico</em> drug-likeness and ADME assessments, including Lipinski's Rule of Five, predicted favorable pharmacokinetic properties. Experimental evaluation confirmed anti-inflammatory efficacy: <em>in vitro</em> protein denaturation assays showed that DDAC display inhibition comparable to diclofenac, while <em>in vivo</em> carrageenan-induced paw edema studies demonstrated significant activity for both derivatives. Collectively, these findings confirm that DDCH and DDAC retain anti-inflammatory properties while offering predicted anticancer potential, with DDAC emerging as the more promising dual-action candidate. This work establishes a rational framework for the further optimization and pharmacological development of diclofenac-based analogues targeting both inflammation and cancer.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"198 ","pages":"Article 111123"},"PeriodicalIF":6.3,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145198245","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}