Computers in biology and medicine最新文献

{"title":"Proximity adjusted centroid mapping for accurate detection of nuclei in dense 3D cell systems.","authors":"Tim Van De Looverbosch, Sarah De Beuckeleer, Frederik De Smet, Jan Sijbers, Winnok H De Vos","doi":"10.1016/j.compbiomed.2024.109561","DOIUrl":"https://doi.org/10.1016/j.compbiomed.2024.109561","url":null,"abstract":"<p><p>In the past decade, deep learning algorithms have surpassed the performance of many conventional image segmentation pipelines. Powerful models are now available for segmenting cells and nuclei in diverse 2D image types, but segmentation in 3D cell systems remains challenging due to the high cell density, the heterogenous resolution and contrast across the image volume, and the difficulty in generating reliable and sufficient ground truth data for model training. Reasoning that most image processing applications rely on nuclear segmentation but do not necessarily require an accurate delineation of their shapes, we implemented Proximity Adjusted Centroid MAPping (PAC-MAP), a 3D U-net based method that predicts the position of nuclear centroids and their proximity to other nuclei. We show that our model outperforms existing methods, predominantly by boosting recall, especially in conditions of high cell density. When trained from scratch with limited expert annotations (30 images), PAC-MAP attained an average F1 score of 0.793 for nuclei centroid prediction in dense spheroids. When pretraining using weakly supervised bulk data (>2300 images) followed by finetuning with the available expert annotations, the average F1 score could be significantly improved to 0.816. We demonstrate the utility of our method for quantifying the absolute cell content of spheroids and comprehensively mapping the infiltration pattern of patient-derived glioblastoma cells in cerebral organoids.</p>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"185 ","pages":"109561"},"PeriodicalIF":7.0,"publicationDate":"2024-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142853367","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":"Developing multifaceted drug synergistic therapeutic strategy against neurological disorders.","authors":"Izza Irum, Fariha Khan, Muhammad Sufyan, Syeda Hafiza Benish Ali, Sidra Rehman","doi":"10.1016/j.compbiomed.2024.109495","DOIUrl":"https://doi.org/10.1016/j.compbiomed.2024.109495","url":null,"abstract":"<p><p>Drug synergism can alter the ultimate biological effects and bioavailability of phytoconstituents. Acetylcholinesterase (AChE) inhibitors as symptomatic drugs are potent therapeutic regimen for neurodegenerative diseases. In this context, this study characterized the synergistic antioxidant, anti-inflammatory and anti-AChE effects of the selected phytochemicals including standard drugs followed by enzyme kinetics, structure-based ligands screening and molecular dynamics simulation study. The synergistic interactions were evaluated through Isoradiation and Synergy finder 3.0 methods. The combinations of Quercetin (QCT), Folic acid (FA), and Swertiamarin (SWT) with specific reference drugs were studied. The combinations of SWT + GA (Gallic acid) and FA + GA at 1:1 (γ:0.10 & 0.08, respectively) showed the significant synergistic antioxidant effect via ABTS assay. Further, in combination, QCT + SWT showed the maximum synergistic effect (γ: 0.02-0.13) in anti-inflammatory assay. Moreover, the combinations QCT, FA, and SWT with reference drug, Donepezil (DP), illustrated potent synergistic activity as anti-AChE in 1:1 proportion (γ: 0.18). The interaction pattern of phytochemicals significantly exhibited synergism (γ < 1) depicting their optimum activity in combinations compared to individual components. Enzyme kinetics evaluation showed the competitive binding of SWT with AChE as of donepezil. All the parameters of ADMET study proposed the QCT and SWT as acceptable oral drug molecules. Computational docking study revealed that QCT and SWT with lowest RMSD (1.096, 2.104) and lowest docking score (-9.831, -7.435 kcal/mol) showed maximum binding efficacy. Furthermore, molecular simulation study depicted the stability of protein-ligand complexes. These findings provide novel insight in the development of dietary treatment based on their synergistic effects for neurological disorders as optimum alternative therapeutic agents.</p>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"185 ","pages":"109495"},"PeriodicalIF":7.0,"publicationDate":"2024-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142853356","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel method to enhance medical image reconstruction using Genetic Algorithm and Incremental Principal Component Analysis.","authors":"Tuğba Özge Onur","doi":"10.1016/j.compbiomed.2024.109527","DOIUrl":"https://doi.org/10.1016/j.compbiomed.2024.109527","url":null,"abstract":"<p><p>Medical imaging has an crucial role in modern healthcare and helps diagnosing and treating for a variety of medical conditions. However, the quality of medical images can be affected by factors such as noise, artifacts, and limited resolution. This paper proposes a novel approach for enhancing the reconstruction of medical images by combining Genetic Algorithm (GA) with Incremental Principal Component Analysis (IPCA). The proposed method aims to improve image quality by extracting relevant features from the original image using GA, followed by reconstruction using IPCA. Through this comprehensive approach, the goal is to enhance the reconstruction of medical images and improve their diagnostic utility in clinical practice. To prove the validity of the proposed method, five different magnetic resonance (MR) images of the shoulder joints are used and the image quality are measured using the signal-to-noise ratio (SNR) terminology with peak signal-to-noise ratio (PSNR), a structural similarity index measure (SSIM) and contrast-to-noise ratio (CNR). The results demonstrate significant improvements in image quality, confirming the effectiveness of the proposed method in enhancing the reconstruction of medical images.</p>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"185 ","pages":"109527"},"PeriodicalIF":7.0,"publicationDate":"2024-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142853351","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":"Novel dilation-erosion labeling technique allows for rapid, accurate and adjustable alignment measurements in primary TKA.","authors":"Aleksander P Mika, Yehyun Suh, Robert W Elrod, Martin Faschingbauer, Daniel C Moyer, J Ryan Martin","doi":"10.1016/j.compbiomed.2024.109571","DOIUrl":"https://doi.org/10.1016/j.compbiomed.2024.109571","url":null,"abstract":"<p><strong>Background: </strong>Optimal implant position and alignment remains a controversial, yet critical topic in primary total knee arthroplasty (TKA). Future study of ideal implant position will require the ability to efficiently measure component positions at scale. Previous algorithms have limited accuracy, do not allow for human oversight and correction in deployment, and require extensive training time and dataset. Therefore, the purpose of this study was to develop and validate a machine learning model that can accurately automate, with surgeon directed adjustment, implant position annotation.</p><p><strong>Methods: </strong>A retrospective series of 295 primary TKAs was identified. The femoral-tibial angle (FTA), distal femoral angle (dFA), and proximal tibial angle (pTA) were manually annotated from the immediate short leg post-op radiograph. We then trained a neural network to predict each annotated landmark using a novel label augmentation procedure of dilation, reweighting, and scheduled erosion steps. The model was compared against diverse models and accuracy was assessed using a validation set of 43 patients and test set of 79 patients.</p><p><strong>Results: </strong>Our proposed model significantly improves accuracy compared to baseline training models across all measures in ten out of eleven cases (p < 1e-22 for each measure). The mean absolute error (difference from manual annotation) was 0.65° for FTA, 1.62° for dFA, and 1.44° for pTA.</p><p><strong>Conclusion: </strong>Utilizing a novel algorithm, trained on a limited dataset, the accuracy of component position was approximately 1.2°. Additionally, the model outputs adjustable points from which the angles are calculated, allowing for clinician oversight and interpretable diagnostics for failure cases.</p>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"185 ","pages":"109571"},"PeriodicalIF":7.0,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142846088","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":"L-MAE: Longitudinal masked auto-encoder with time and severity-aware encoding for diabetic retinopathy progression prediction.","authors":"Rachid Zeghlache, Pierre-Henri Conze, Mostafa El Habib Daho, Yihao Li, Alireza Rezaei, Hugo Le Boité, Ramin Tadayoni, Pascal Massin, Béatrice Cochener, Ikram Brahim, Gwenolé Quellec, Mathieu Lamard","doi":"10.1016/j.compbiomed.2024.109508","DOIUrl":"https://doi.org/10.1016/j.compbiomed.2024.109508","url":null,"abstract":"<p><p>Pre-training strategies based on self-supervised learning (SSL) have demonstrated success as pretext tasks for downstream tasks in computer vision. However, while SSL methods are often domain-agnostic, their direct application to medical imaging is challenging due to the distinct nature of medical images, including specific anatomical and temporal patterns relevant to disease progression. Additionally, traditional SSL pretext tasks often lack the contextual knowledge that is essential for clinical decision support. In this paper, we developed a longitudinal masked auto-encoder (MAE) that builds on the Transformer-based MAE architecture, specifically introducing a time-aware position embedding and a disease progression-aware masking strategy. Unlike traditional sequential approaches, our method incorporates the actual time intervals between examinations, allowing for better capture of temporal trends. Furthermore, the masking strategy evolves in alignment with disease progression during follow-up exams to capture pathological changes, improving disease progression assessments. Using the OPHDIAT dataset, a large-scale longitudinal screening dataset for diabetic retinopathy (DR), we evaluated our pre-trained model by predicting the severity level at the next visit within three years, based on past examination series. Our findings demonstrate that both the time-aware position embedding and the disease progression-informed masking significantly enhance predictive accuracy. Compared to conventional baseline models and standard longitudinal Transformers, these simple yet effective adaptations substantially improve the predictive power of deep classification models in this domain.</p>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"185 ","pages":"109508"},"PeriodicalIF":7.0,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142846082","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 an integrated imaging for melanoma diagnosis: A review of multispectral, hyperspectral, and thermal technologies with preliminary system development.","authors":"Maria Oniga, Alina Sultana, Bogdan Alexandrescu, Olguța Orzan","doi":"10.1016/j.compbiomed.2024.109570","DOIUrl":"https://doi.org/10.1016/j.compbiomed.2024.109570","url":null,"abstract":"<p><p>The diagnosis of melanoma traditionally relies on visual inspection or on the use of the dermoscope, which do not have capabilities for early and precise detection. In this review, we aimed to explore other imaging technologies that can provide non-invasive and detailed information on skin lesions, such as multispectral, hyperspectral and thermal imaging. In this regard, the systems were evaluated in terms of hardware, performance and clinical applications. Since there is currently a very big interest in developing artificial intelligence (AI) applications in dermatology, the review also focuses on analysing studies that integrated this technology with newer imaging systems. To obtain clinical validation for such systems, there is an extensive need for publicly available datasets, as the current ones are limited. Expanding and obtaining new datasets is crucial in advancing research for a more accurate melanoma diagnosis. Taking into consideration the benefits that these imaging modalities can provide if they are combined with AI, we propose a prototype that can distinguish between melanoma and its precursor, the nevus, for which the set-up, components, imaging processing pipeline and classification techniques are described. The final system benefits of the advantages provided by near infrared, thermal and visible cameras, that allow a more in-depth characterizations of melanoma for a better understanding of its behaviour, an early detection improvement and diagnostic precision.</p>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"185 ","pages":"109570"},"PeriodicalIF":7.0,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142846096","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":"Simulated effects of surgical corrections on bone-implant micromotion and implant stresses in paediatric proximal femoral osteotomy.","authors":"Alireza Y Bavil, Emmanuel Eghan-Acquah, Ayda Karimi Dastgerdi, Laura E Diamond, Rod Barrett, Henry Pj Walsh, Martina Barzan, David J Saxby, Stefanie Feih, Christopher P Carty","doi":"10.1016/j.compbiomed.2024.109544","DOIUrl":"https://doi.org/10.1016/j.compbiomed.2024.109544","url":null,"abstract":"<p><strong>Background and objective: </strong>Proximal femoral osteotomy (PFO) is a surgical intervention, typically performed on paediatric population, that aims to correct femoral deformities caused by different pathologies (e.g., slipped capital femoral epiphysis). A PFO involves introduction of an implant to fix the proximal and distal sections of femur following the surgical corrections. The femoral neck-shaft angle (NSA) and anteversion angle (AVA) are key geometric parameters that influence PFO outcomes. To date, the effects of NSA and AVA on bone-implant system mechanics in paediatric populations have not been examined.</p><p><strong>Methods: </strong>This study used an established neuromusculoskeletal modelling process paired with finite element analysis to determine the sensitivity of the implanted femur's mechanics to variations in NSA and AVA during the stance phase of walking. Three male patients aged 9-12 years with different pathology (Spastic diplegia, Perthes disease and Slipped Capital Femoral Epiphysis), weight (377, 747, 842 N), height (1.39, 1.55, 1.71 m) and femur lengths (34.1, 39.4, 43.7 cm) and geometries (NSA: 143, 102, 111 deg; AVA: 29, 17, -22 deg) were examined. For each patient, a three-dimensional bone model was created from computed tomography imaging and digital surgical corrections were applied to systematically vary the NSA and AVA. Personalized motion and loading conditions driven from a neuromusculoskeletal modelling process were applied to each model and its associated permutations of NSA and AVA.</p><p><strong>Results: </strong>Results indicated significant intra-participant variability in post-PFO bone-implant micromotion and peak von Mises stress on implant. For models with a post-surgery NSA of 135° and AVA of 12°, the averaged micromotion increased by 87 % and the peak von Mises stress decreased by 63% between patient 1 and 2. Between patient 2 and 3, the averaged micromotion decreased by 55% while the peak von Mises stress increased by 84%.</p><p><strong>Conclusions: </strong>Furthermore, post-PFO bone-implant mechanics were sensitive to variation in NSA and AVA in a subject-specific manner. Optimization of PFO planning is recommended based on patient-specific characteristics.</p>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"185 ","pages":"109544"},"PeriodicalIF":7.0,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142846092","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 an interpretable breast cancer detection and diagnosis system.","authors":"Cristiana Moroz-Dubenco, Adél Bajcsi, Anca Andreica, Camelia Chira","doi":"10.1016/j.compbiomed.2024.109520","DOIUrl":"https://doi.org/10.1016/j.compbiomed.2024.109520","url":null,"abstract":"<p><p>According to the World Health Organization, breast cancer becomes fatal only if it spreads throughout the body. Therefore, regular screening is essential. Whilst mammography is the most frequently used technique, its interpretation can be challenging and time-consuming. For this reason, computer-aided detection and diagnosis systems are increasingly being used for second opinion. However, in order for doctors to trust such systems, they need to understand their decisions. We propose an automated and interpretable system for the detection and diagnosis of breast cancer, encompassing five steps. After a robust pre-processing and an unsupervised segmentation, we analyze four feature extraction techniques, both textural and shape-based, and three methods for feature selection. To facilitate interpretation, we employ the Decision Tree algorithm for benign/malignant classification and experiment with different methods to avoid overfitting: pre-pruning, post-pruning, and ensemble-based (Random Forest classifier). Our system reaches a maximum accuracy of 95% and 100% precision and specificity when tested on images from the mini-MIAS dataset, while also offering its users the possibility to analyze each of the steps.</p>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"185 ","pages":"109520"},"PeriodicalIF":7.0,"publicationDate":"2024-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142845526","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":"Subspace learning using low-rank latent representation learning and perturbation theorem: Unsupervised gene selection.","authors":"Amir Moslemi, Fariborz Baghaei Naeini","doi":"10.1016/j.compbiomed.2024.109567","DOIUrl":"https://doi.org/10.1016/j.compbiomed.2024.109567","url":null,"abstract":"<p><p>In recent years, gene expression data analysis has gained growing significance in the fields of machine learning and computational biology. Typically, microarray gene datasets exhibit a scenario where the number of features exceeds the number of samples, resulting in an ill-posed and underdetermined equation system. The presence of redundant features in high-dimensional data leads to suboptimal performance and increased computational time for learning algorithms. Although feature extraction and feature selection are two approaches that can be employed to deal with this challenge, feature selection has greater interpretability ability which causes it to receive more attention. In this study, we propose an unsupervised feature selection which is based on pseudo label latent representation learning and perturbation theory. In the first step, pseudo labels are extracted and constructed using latent representation learning. In the second step, the least square problem is solved for original data matrix and perturbed data matrix. Features are clustered based on the similarity between the original data matrix and the perturbed data matrix using k-means. In the last step, features in each subcluster are ranked based on information gain criterion. To showcase the efficacy of the proposed approach, numerical experiments were carried out on six benchmark microarray datasets and two RNA-Sequencing benchmark datasets. The outcomes indicate that the proposed technique surpasses eight state-of-the-art unsupervised feature selection methods in both clustering accuracy and normalized mutual information.</p>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"185 ","pages":"109567"},"PeriodicalIF":7.0,"publicationDate":"2024-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142827619","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 evaluation of interactive dynamics for a full transcatheter aortic valve device in a patient-specific aortic root.","authors":"Jingwen Zhang, Ran He, Jia Wu, Zhihao Fan, Dong Liu, Andy Gleadall, Liguo Zhao, Simin Li","doi":"10.1016/j.compbiomed.2024.109512","DOIUrl":"https://doi.org/10.1016/j.compbiomed.2024.109512","url":null,"abstract":"<p><p>Transcatheter aortic valve implantation (TAVI) has become a key treatment for severe aortic stenosis, especially for patients unsuitable for surgery. Since its introduction in 2002, TAVI has advanced significantly due to improvements in imaging, operator skills, and device engineering. Despite these innovations, challenges in device sizing and positioning remain, complicating outcome predictions. Computational modelling is a powerful tool to aid TAVI device design and to understand its interactive behaviour with the aortic root during the deployment. Previous studies often simplified tissue properties, neglected patient-specific geometries or omitted crucial elements such as leaflets and fabric. This paper presents a numerical framework capable of simulating the whole crimping and deployment process of a full TAVI device in a patient-specific aortic root including the native leaflets and calcifications. We conduct a comprehensive investigation into the mechanical behaviour of the TAVI and its interactions with patient-specific aortic root through dynamic finite element analysis during the deployment process, with validation against experimental results. Additionally, we examined the influence of applied pressure during balloon inflation on the interactive dynamics of the entire model. The study concludes that selecting optimal balloon pressures is crucial for enhancing TAVI device performance and reducing complications. Numerical simulations demonstrate that appropriate balloon pressure ensures sufficient flow area and effective contact pressure between the TAVI and the aortic root, while minimising deformation and the risk of paravalvular leak.</p>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"185 ","pages":"109512"},"PeriodicalIF":7.0,"publicationDate":"2024-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142827617","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}