Computer methods and programs in biomedicine最新文献_第8页

Digital pathology-based prognostic model for hepatocellular carcinoma: Integrating pathomics signatures with clinical parameters for recurrence prediction and biological interpretation 基于病理的肝细胞癌数字预后模型：将病理特征与复发预测和生物学解释的临床参数相结合。

IF 4.8 2区医学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computer methods and programs in biomedicine

Pub Date : 2026-02-01 Epub Date: 2025-11-27 DOI: 10.1016/j.cmpb.2025.109180

Qi Wang , Yuxi Huang , Yu Zhang , Yu Zhu , Peng Hu , Yongfu Xu , Zhen-yu Jiang , Long Liu , Shao-wei Li

Background

Hepatocellular carcinoma (HCC) remains a therapeutic challenge due to high post-resection recurrence rates and heterogeneous outcomes. We developed and validated a digital pathology-based prognostic model combining pathomics signatures with clinical parameters to predict recurrence and elucidate biological mechanisms.

Methods

In this multicenter retrospective study, 294 HCC patients (training set: n = 198; validation set: n = 96) undergoing curative hepatectomy were analyzed. Pathomics features were quantitatively extracted from H&E-stained whole-slide images. Predictive modeling incorporated machine learning approaches (DT, KNN, LASSO, NB, RF, SVM) with clinical variables. Model performance was evaluated through ROC analysis, calibration, and decision curve analysis. Biological interpretation leveraged TCGA transcriptomic data analyzed via GSEA and WGCNA.

Results

Tumor and peri‑tumor pathomics parameters showed some complementarity in the prediction of HCC recurrence. The combined LASSO-based model showed the best predictive efficacy, with AUCs of 0.850 and 0.807 in the training and validation sets, respectively. The integrated pathomics-clinical model achieved AUCs of 0.893 and 0.860 in training and validation sets. Bioinformatics analysis suggested that the pathomics was correlated with the tumor immune microenvironment, as verified by multiple immunofluorescence staining of the validation set.

Conclusion

This study establishes a robust digital pathology framework that not only improves HCC recurrence prediction beyond conventional biomarkers but also provides mechanistic insights into tumor-immune crosstalk.

背景：肝细胞癌（HCC）仍然是一个治疗挑战，由于高术后复发率和异质性的结果。我们开发并验证了一种基于病理的数字预后模型，该模型结合了病理特征和临床参数来预测复发并阐明生物学机制。方法：在这项多中心回顾性研究中，对294例行根治性肝切除术的HCC患者（训练组198例，验证组96例）进行分析。从h&e染色的全片图像中定量提取病理特征。预测建模结合了机器学习方法（DT、KNN、LASSO、NB、RF、SVM）和临床变量。通过ROC分析、校正和决策曲线分析来评估模型的性能。生物解释利用了通过GSEA和WGCNA分析的TCGA转录组数据。结果：肿瘤和肿瘤周围病理参数在预测HCC复发方面具有一定的互补性。基于lasso的联合模型预测效果最好，训练集和验证集的auc分别为0.850和0.807。综合病理-临床模型在训练集和验证集的auc分别为0.893和0.860。生物信息学分析表明，病理与肿瘤免疫微环境相关，验证组多次免疫荧光染色证实。结论：该研究建立了一个强大的数字病理学框架，不仅可以提高HCC复发预测，而且可以提供肿瘤免疫串扰的机制见解。

{"title":"Digital pathology-based prognostic model for hepatocellular carcinoma: Integrating pathomics signatures with clinical parameters for recurrence prediction and biological interpretation","authors":"Qi Wang , Yuxi Huang , Yu Zhang , Yu Zhu , Peng Hu , Yongfu Xu , Zhen-yu Jiang , Long Liu , Shao-wei Li","doi":"10.1016/j.cmpb.2025.109180","DOIUrl":"10.1016/j.cmpb.2025.109180","url":null,"abstract":"<div><h3>Background</h3><div>Hepatocellular carcinoma (HCC) remains a therapeutic challenge due to high post-resection recurrence rates and heterogeneous outcomes. We developed and validated a digital pathology-based prognostic model combining pathomics signatures with clinical parameters to predict recurrence and elucidate biological mechanisms.</div></div><div><h3>Methods</h3><div>In this multicenter retrospective study, 294 HCC patients (training set: <em>n</em> = 198; validation set: <em>n</em> = 96) undergoing curative hepatectomy were analyzed. Pathomics features were quantitatively extracted from H&E-stained whole-slide images. Predictive modeling incorporated machine learning approaches (DT, KNN, LASSO, NB, RF, SVM) with clinical variables. Model performance was evaluated through ROC analysis, calibration, and decision curve analysis. Biological interpretation leveraged TCGA transcriptomic data analyzed via GSEA and WGCNA.</div></div><div><h3>Results</h3><div>Tumor and peri‑tumor pathomics parameters showed some complementarity in the prediction of HCC recurrence. The combined LASSO-based model showed the best predictive efficacy, with AUCs of 0.850 and 0.807 in the training and validation sets, respectively. The integrated pathomics-clinical model achieved AUCs of 0.893 and 0.860 in training and validation sets. Bioinformatics analysis suggested that the pathomics was correlated with the tumor immune microenvironment, as verified by multiple immunofluorescence staining of the validation set.</div></div><div><h3>Conclusion</h3><div>This study establishes a robust digital pathology framework that not only improves HCC recurrence prediction beyond conventional biomarkers but also provides mechanistic insights into tumor-immune crosstalk.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"275 ","pages":"Article 109180"},"PeriodicalIF":4.8,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145676357","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}

引用次数: 0

DoubleBlock-ViT: A MaxViT-based enhancement and dual-path skip connections for brain tumor segmentation in MRI scans DoubleBlock-ViT：一种基于maxvit的增强和双路径跳过连接在MRI扫描中用于脑肿瘤分割

IF 4.8 2区医学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computer methods and programs in biomedicine

Pub Date : 2026-02-01 Epub Date: 2025-11-17 DOI: 10.1016/j.cmpb.2025.109165

Thien B. Nguyen-Tat, Lap Quang Truong, Khanh Quoc Truong, Tai Huy Ngo

Background:

Manual interpretation of brain tumor regions in MRI scans demands substantial medical expertise, is time-consuming, and is prone to human error, especially when radiologists must review large scan volumes, leading to fatigue and reduced diagnostic accuracy.

Objective:

This study aims to propose a lightweight segmentation framework with strong global contextual awareness, enabling accurate and efficient delineation of tumor regions for radiological analysis and downstream clinical applications.

Methods:

This study leverages multimodal MRI inputs (T1, T1Gd, T2, T2FLAIR) and introduces a novel hybrid 3D U-Net architecture that integrates Convolutional Neural Networks (CNNs) with Transformer-based modeling. The architecture incorporates a 3D DoubleBlock-ViT Transformer encoder to capture long-range dependencies and global context via attention mechanisms, while a Dual-Path Fusion Block with CNN-based skip connections preserves fine-grained spatial details and enhances feature transfer.

Results:

Evaluations on BraTS2020 and BraTS2021 benchmark datasets yield Dice scores of 80.11% (ET), 86.60% (TC), and 91.20% (WT) on BraTS2020, and 87.82% (ET), 91.61% (TC), and 92.31% (WT) on BraTS2021, outperforming several state-of-the-art methods.

Conclusions:

The proposed model delivers state-of-the-art accuracy with only 7.8 million parameters and low computational demands, making it well-suited for clinical deployment and integration into radiomics pipelines for precision oncology. By leveraging multimodal MRI inputs and advanced feature extraction mechanisms, our approach directly aligns with current advances in AI-driven radiomics. The codes and trained models will be publicly available at https://github.com/Laptq201/DoubleBlock-ViT-Unet-segment.

背景：在MRI扫描中手动解释脑肿瘤区域需要大量的医学专业知识，耗时，并且容易出现人为错误，特别是当放射科医生必须检查大量扫描时，导致疲劳和诊断准确性降低。目的：本研究旨在提出一种具有强大全局上下文意识的轻量级分割框架，为放射学分析和下游临床应用提供准确有效的肿瘤区域描绘。方法：本研究利用多模态MRI输入（T1, T1Gd, T2, T2FLAIR），并引入了一种新的混合3D U-Net架构，该架构将卷积神经网络（cnn）与基于变压器的建模集成在一起。该架构结合了3D DoubleBlock-ViT Transformer编码器，通过注意机制捕获远程依赖关系和全局上下文，而基于cnn的跳过连接的双路径融合块保留了细粒度的空间细节并增强了特征传递。结果：对BraTS2020和BraTS2021基准数据集进行评估，在BraTS2020上的Dice得分为80.11% （ET）、86.60% （TC）和91.20% (WT)，在BraTS2021上的Dice得分为87.82% （ET）、91.61% （TC）和92.31% (WT)，优于几种最先进的方法。结论：该模型仅需要780万个参数，计算量低，具有最先进的精度，非常适合临床部署和整合到精确肿瘤学的放射组学管道中。通过利用多模态MRI输入和先进的特征提取机制，我们的方法直接与人工智能驱动的放射组学的当前进展保持一致。代码和经过训练的模型将在https://github.com/Laptq201/DoubleBlock-ViT-Unet-segment上公开提供。

{"title":"DoubleBlock-ViT: A MaxViT-based enhancement and dual-path skip connections for brain tumor segmentation in MRI scans","authors":"Thien B. Nguyen-Tat, Lap Quang Truong, Khanh Quoc Truong, Tai Huy Ngo","doi":"10.1016/j.cmpb.2025.109165","DOIUrl":"10.1016/j.cmpb.2025.109165","url":null,"abstract":"<div><h3>Background:</h3><div>Manual interpretation of brain tumor regions in MRI scans demands substantial medical expertise, is time-consuming, and is prone to human error, especially when radiologists must review large scan volumes, leading to fatigue and reduced diagnostic accuracy.</div></div><div><h3>Objective:</h3><div>This study aims to propose a lightweight segmentation framework with strong global contextual awareness, enabling accurate and efficient delineation of tumor regions for radiological analysis and downstream clinical applications.</div></div><div><h3>Methods:</h3><div>This study leverages multimodal MRI inputs (T1, T1Gd, T2, T2FLAIR) and introduces a novel hybrid 3D U-Net architecture that integrates Convolutional Neural Networks (CNNs) with Transformer-based modeling. The architecture incorporates a 3D DoubleBlock-ViT Transformer encoder to capture long-range dependencies and global context via attention mechanisms, while a Dual-Path Fusion Block with CNN-based skip connections preserves fine-grained spatial details and enhances feature transfer.</div></div><div><h3>Results:</h3><div>Evaluations on BraTS2020 and BraTS2021 benchmark datasets yield Dice scores of 80.11% (ET), 86.60% (TC), and 91.20% (WT) on BraTS2020, and 87.82% (ET), 91.61% (TC), and 92.31% (WT) on BraTS2021, outperforming several state-of-the-art methods.</div></div><div><h3>Conclusions:</h3><div>The proposed model delivers state-of-the-art accuracy with only 7.8 million parameters and low computational demands, making it well-suited for clinical deployment and integration into radiomics pipelines for precision oncology. By leveraging multimodal MRI inputs and advanced feature extraction mechanisms, our approach directly aligns with current advances in AI-driven radiomics. The codes and trained models will be publicly available at <span><span>https://github.com/Laptq201/DoubleBlock-ViT-Unet-segment</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"274 ","pages":"Article 109165"},"PeriodicalIF":4.8,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145569489","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}

引用次数: 0

A High-resolution dataset for AI-driven segmentation and analysis of drug-treated breast tumor spheroids 用于人工智能驱动的药物治疗乳腺肿瘤球体分割和分析的高分辨率数据集

IF 4.8 2区医学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computer methods and programs in biomedicine

Pub Date : 2026-02-01 Epub Date: 2025-11-04 DOI: 10.1016/j.cmpb.2025.109141

Amir Tahmasbi , Akram Ahvaraki , Ebrahim Behroodi , Aboozar Ghaffari , Zeinab Bagheri , Faezeh Vakhshiteh , Hamid Latifi , Zahra Madjd

Background and objective

Three-dimensional (3D) tumor spheroids are widely adopted in preclinical drug screening for their ability to mimic the complexity of in vivo tumor microenvironments. Nevertheless, the high-throughput analysis of such models, especially for quantifying drug responses, remains a significant challenge. This study aims to introduce a high-resolution, publicly available dataset to facilitate AI-driven segmentation and analysis of drug-treated breast tumor spheroids.

Methods

Heterotypic spheroids consisting of MDA-MB-231 breast cancer cells and human fibroblasts were cultured in a microfluidic chip and subjected to either treatment with liposomal doxorubicin or left untreated. Microscopic imaging was conducted over eight consecutive days, resulting in 95 high-resolution images. These were preprocessed and divided into 2980 image tiles (512 × 512 pixels), followed by semi-automated annotation. The dataset was evaluated using three deep learning segmentation models: U-Net, Fully Convolutional Network (FCN), Mask R-CNN, YOLOv12-Seg, and DeepLab. Morphological features extracted from the segmented spheroids were analyzed using both statistical and machine learning techniques.

Results

Among the models tested, DeepLab achieved the highest segmentation accuracy with a Jaccard index of 91.17 %. Key morphological descriptors—area, perimeter, inradius, and boundary complexity—were extracted and analyzed using Generalized Estimating Equations, revealing statistically significant differences (p < 0.05) between control and treated spheroids. Classification using a Support Vector Machine trained on features reduced via Principal Component Analysis resulted in 96 % accuracy in distinguishing the two groups.

Conclusions

The HTS-Seg dataset provides a high-quality image resource with corresponding annotations and morphological features, supporting the development and validation of segmentation and classification models in biomedical image analysis. This work enables more accurate in vitro evaluation of drug effects on 3D tumor spheroid models and contributes to advancements in AI-assisted cancer research.

背景与目的三维肿瘤球体由于能够模拟体内肿瘤微环境的复杂性，在临床前药物筛选中被广泛采用。然而，这种模型的高通量分析，特别是用于定量药物反应，仍然是一个重大挑战。本研究旨在引入一个高分辨率、公开可用的数据集，以促进人工智能驱动的药物治疗乳腺肿瘤球体的分割和分析。方法在微流控芯片中培养由MDA-MB-231乳腺癌细胞和人成纤维细胞组成的异型球体，并用阿霉素脂质体处理或不处理。显微成像连续进行了8天，获得了95张高分辨率图像。对这些图像进行预处理并划分为2980个图像块（512 × 512像素），然后进行半自动注释。使用三种深度学习分割模型对数据集进行评估：U-Net、全卷积网络（FCN）、Mask R-CNN、YOLOv12-Seg和DeepLab。使用统计和机器学习技术对从分割球体中提取的形态学特征进行分析。结果DeepLab模型分割准确率最高，Jaccard指数为91.17%。关键形态描述符——面积、周长、内半径和边界复杂性——被提取并使用广义估计方程进行分析，结果显示对照组和处理过的球体之间存在统计学显著差异（p < 0.05）。使用通过主成分分析减少特征训练的支持向量机进行分类，区分两组的准确率为96%。结论HTS-Seg数据集提供了高质量的图像资源，具有相应的注释和形态学特征，支持生物医学图像分析中分割和分类模型的开发和验证。这项工作能够更准确地在体外评估药物对3D肿瘤球体模型的影响，并有助于人工智能辅助癌症研究的进步。

{"title":"A High-resolution dataset for AI-driven segmentation and analysis of drug-treated breast tumor spheroids","authors":"Amir Tahmasbi , Akram Ahvaraki , Ebrahim Behroodi , Aboozar Ghaffari , Zeinab Bagheri , Faezeh Vakhshiteh , Hamid Latifi , Zahra Madjd","doi":"10.1016/j.cmpb.2025.109141","DOIUrl":"10.1016/j.cmpb.2025.109141","url":null,"abstract":"<div><h3>Background and objective</h3><div>Three-dimensional (3D) tumor spheroids are widely adopted in preclinical drug screening for their ability to mimic the complexity of in vivo tumor microenvironments. Nevertheless, the high-throughput analysis of such models, especially for quantifying drug responses, remains a significant challenge. This study aims to introduce a high-resolution, publicly available dataset to facilitate AI-driven segmentation and analysis of drug-treated breast tumor spheroids.</div></div><div><h3>Methods</h3><div>Heterotypic spheroids consisting of MDA-MB-231 breast cancer cells and human fibroblasts were cultured in a microfluidic chip and subjected to either treatment with liposomal doxorubicin or left untreated. Microscopic imaging was conducted over eight consecutive days, resulting in 95 high-resolution images. These were preprocessed and divided into 2980 image tiles (512 × 512 pixels), followed by semi-automated annotation. The dataset was evaluated using three deep learning segmentation models: U-Net, Fully Convolutional Network (FCN), Mask R-CNN, YOLOv12-Seg, and DeepLab. Morphological features extracted from the segmented spheroids were analyzed using both statistical and machine learning techniques.</div></div><div><h3>Results</h3><div>Among the models tested, DeepLab achieved the highest segmentation accuracy with a Jaccard index of 91.17 %. Key morphological descriptors—area, perimeter, inradius, and boundary complexity—were extracted and analyzed using Generalized Estimating Equations, revealing statistically significant differences (<em>p</em> < 0.05) between control and treated spheroids. Classification using a Support Vector Machine trained on features reduced via Principal Component Analysis resulted in 96 % accuracy in distinguishing the two groups.</div></div><div><h3>Conclusions</h3><div>The HTS-Seg dataset provides a high-quality image resource with corresponding annotations and morphological features, supporting the development and validation of segmentation and classification models in biomedical image analysis. This work enables more accurate in vitro evaluation of drug effects on 3D tumor spheroid models and contributes to advancements in AI-assisted cancer research.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"274 ","pages":"Article 109141"},"PeriodicalIF":4.8,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145464460","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}

引用次数: 0

Renal Cell Carcinoma subtyping: Learning from multi-resolution localization 肾细胞癌亚型：从多分辨率定位中学习

IF 4.8 2区医学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computer methods and programs in biomedicine

Pub Date : 2026-02-01 Epub Date: 2025-11-07 DOI: 10.1016/j.cmpb.2025.109155

Mohamad Mohamad , Francesco Ponzio , Santa Di Cataldo , Damien Ambrosetti , Xavier Descombes

Background and Objective

Renal Cell Carcinoma (RCC) is often diagnosed at advanced stages, limiting treatment options. Since prognosis depends on tumour subtype, accurate and efficient classification is essential. Artificial intelligence tools can assist diagnosis, yet their dependence on large annotated datasets hinders broader adoption. This study investigates a Self-Supervised Learning (SSL) framework that exploits the multi-resolution structure of Whole histological Slide Images (WSIs) to reduce annotation requirements while maintaining reliable diagnostic performance.

Methods:

We developed a SSL model inspired by the pathologist’s multi-scale reasoning, integrating information across magnification levels. Robustness and generalization were evaluated through an external validation on a public RCC benchmark and one internal validation using cohorts from the same institution but collected in different periods, with distinct scanners and laboratory workflows.

Results and Conclusions

The proposed SSL approach demonstrated stable classification performance across all validation settings, reducing dependence on manual labels and improving robustness under heterogeneous acquisition conditions. These findings support its potential as a generalizable and annotation-efficient strategy for RCC subtype classification.

背景和目的肾细胞癌（RCC）通常在晚期被诊断出来，限制了治疗选择。由于预后取决于肿瘤亚型，准确和有效的分类是必不可少的。人工智能工具可以帮助诊断，但它们对大型注释数据集的依赖阻碍了更广泛的采用。本研究探讨了一种自我监督学习（SSL）框架，该框架利用全组织学幻灯片图像（wsi）的多分辨率结构来减少注释要求，同时保持可靠的诊断性能。方法：我们开发了一个受病理学家多尺度推理启发的SSL模型，整合了放大水平上的信息。鲁棒性和泛化性通过公共RCC基准的外部验证和来自同一机构但在不同时期收集的队列的内部验证来评估，使用不同的扫描仪和实验室工作流程。结果和结论提出的SSL方法在所有验证设置下都表现出稳定的分类性能，减少了对手动标签的依赖，提高了异构获取条件下的鲁棒性。这些发现支持了它作为RCC亚型分类的一种可推广且注释有效的策略的潜力。

{"title":"Renal Cell Carcinoma subtyping: Learning from multi-resolution localization","authors":"Mohamad Mohamad , Francesco Ponzio , Santa Di Cataldo , Damien Ambrosetti , Xavier Descombes","doi":"10.1016/j.cmpb.2025.109155","DOIUrl":"10.1016/j.cmpb.2025.109155","url":null,"abstract":"<div><h3>Background and Objective</h3><div>Renal Cell Carcinoma (RCC) is often diagnosed at advanced stages, limiting treatment options. Since prognosis depends on tumour subtype, accurate and efficient classification is essential. Artificial intelligence tools can assist diagnosis, yet their dependence on large annotated datasets hinders broader adoption. This study investigates a Self-Supervised Learning (SSL) framework that exploits the multi-resolution structure of Whole histological Slide Images (WSIs) to reduce annotation requirements while maintaining reliable diagnostic performance.</div></div><div><h3>Methods:</h3><div>We developed a SSL model inspired by the pathologist’s multi-scale reasoning, integrating information across magnification levels. Robustness and generalization were evaluated through an external validation on a public RCC benchmark and one internal validation using cohorts from the same institution but collected in different periods, with distinct scanners and laboratory workflows.</div></div><div><h3>Results and Conclusions</h3><div>The proposed SSL approach demonstrated stable classification performance across all validation settings, reducing dependence on manual labels and improving robustness under heterogeneous acquisition conditions. These findings support its potential as a generalizable and annotation-efficient strategy for RCC subtype classification.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"274 ","pages":"Article 109155"},"PeriodicalIF":4.8,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145464521","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}

引用次数: 0

Benchmarking parametric models of disease progression for early detection of cognitive decline 标杆参数模型的疾病进展的早期检测认知能力下降

IF 4.8 2区医学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computer methods and programs in biomedicine

Pub Date : 2026-02-01 Epub Date: 2025-11-11 DOI: 10.1016/j.cmpb.2025.109162

Carlos Platero , Jorge Bengoa

Background and Objective:

Disease progression models (DPMs) are valuable tools for characterizing early cognitive decline in Alzheimer’s Disease (AD) and supporting clinical decision-making. This study aimed to (1) evaluate the diagnostic and prognostic performance of parametric DPMs, (2) identify optimal subsets of neuropsychological markers for DPM construction, and (3) benchmark three parametric DPM frameworks in early detection tasks.

Methods:

We analyzed longitudinal neuropsychological data from 1163 participants classified as cognitively unimpaired (CU) or with mild cognitive impairment (MCI). Three DPM approaches (Leaspy, RPDPM, and GRACE) were trained on selected marker subsets and evaluated using metrics related to diagnostic accuracy, time to conversion estimation, and robustness to missing data. Model performance was assessed via detection rates, area under the curve (AUC), mean absolute error (MAE), and Pearson correlation between estimated/observed onset ages.

Results:

Leaspy achieved the highest diagnostic accuracy with an AUC of 0.96 and strong correlation with observed conversion time (r = 0.78). RPDPM showed superior robustness to missing data and maintained accurate predictions even with up to 40% data loss. GRACE offered the best trajectory fit (lowest error) but lower sensitivity to clinical transitions. A compact combination of neuropsychological tests, particularly CDRSB, ADAS13, and MMSE, was sufficient for reliable model training. Prognostic evaluation demonstrated that Leaspy provided the most consistent identification of individuals who converted to mild cognitive impairment within five years.

Conclusions:

Parametric DPMs based solely on neuropsychological measures can effectively support early detection and prognosis of cognitive decline. Leaspy showed the best overall performance, while RPDPM proved more resilient to missing data. These models enable individualized disease timelines and can inform clinical decision-making and patient stratification. All code and data used are publicly available, facilitating reproducibility and clinical translation.

背景与目的：疾病进展模型（dpm）是表征阿尔茨海默病（AD）早期认知能力下降和支持临床决策的宝贵工具。本研究旨在(1)评估参数DPM的诊断和预后性能，(2)确定构建DPM的最佳神经心理学标记子集，以及(3)在早期检测任务中对三种参数DPM框架进行基准测试。方法：我们分析了1163名认知未受损（CU）或轻度认知障碍（MCI）参与者的纵向神经心理学数据。三种DPM方法（Leaspy， RPDPM和GRACE）在选定的标记子集上进行训练，并使用与诊断准确性，转换估计时间和缺失数据的鲁棒性相关的指标进行评估。通过检出率、曲线下面积（AUC）、平均绝对误差（MAE）和估计/观察发病年龄之间的Pearson相关性来评估模型的性能。结果：Leaspy的诊断准确率最高，AUC为0.96，与观察到的转换时间有很强的相关性（r = 0.78）。RPDPM对丢失的数据表现出优异的鲁棒性，即使在高达40%的数据丢失情况下也能保持准确的预测。GRACE提供了最佳的轨迹拟合（最低误差），但对临床转变的敏感性较低。神经心理学测试的紧密结合，特别是CDRSB、ADAS13和MMSE，足以进行可靠的模型训练。预后评估表明，Leaspy提供了最一致的识别个体谁转化为轻度认知障碍在五年内。结论：单纯基于神经心理学测量的参数化dpm能有效支持认知衰退的早期发现和预后。Leaspy表现出最好的整体性能，而RPDPM对丢失数据的适应性更强。这些模型可以实现个性化的疾病时间表，并可以为临床决策和患者分层提供信息。所有使用的代码和数据都是公开的，促进了可重复性和临床翻译。

{"title":"Benchmarking parametric models of disease progression for early detection of cognitive decline","authors":"Carlos Platero , Jorge Bengoa","doi":"10.1016/j.cmpb.2025.109162","DOIUrl":"10.1016/j.cmpb.2025.109162","url":null,"abstract":"<div><h3>Background and Objective:</h3><div>Disease progression models (DPMs) are valuable tools for characterizing early cognitive decline in Alzheimer’s Disease (AD) and supporting clinical decision-making. This study aimed to (1) evaluate the diagnostic and prognostic performance of parametric DPMs, (2) identify optimal subsets of neuropsychological markers for DPM construction, and (3) benchmark three parametric DPM frameworks in early detection tasks.</div></div><div><h3>Methods:</h3><div>We analyzed longitudinal neuropsychological data from 1163 participants classified as cognitively unimpaired (CU) or with mild cognitive impairment (MCI). Three DPM approaches (Leaspy, RPDPM, and GRACE) were trained on selected marker subsets and evaluated using metrics related to diagnostic accuracy, time to conversion estimation, and robustness to missing data. Model performance was assessed via detection rates, area under the curve (AUC), mean absolute error (MAE), and Pearson correlation between estimated/observed onset ages.</div></div><div><h3>Results:</h3><div>Leaspy achieved the highest diagnostic accuracy with an AUC of 0.96 and strong correlation with observed conversion time (r = 0.78). RPDPM showed superior robustness to missing data and maintained accurate predictions even with up to 40% data loss. GRACE offered the best trajectory fit (lowest error) but lower sensitivity to clinical transitions. A compact combination of neuropsychological tests, particularly CDRSB, ADAS13, and MMSE, was sufficient for reliable model training. Prognostic evaluation demonstrated that Leaspy provided the most consistent identification of individuals who converted to mild cognitive impairment within five years.</div></div><div><h3>Conclusions:</h3><div>Parametric DPMs based solely on neuropsychological measures can effectively support early detection and prognosis of cognitive decline. Leaspy showed the best overall performance, while RPDPM proved more resilient to missing data. These models enable individualized disease timelines and can inform clinical decision-making and patient stratification. All code and data used are publicly available, facilitating reproducibility and clinical translation.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"274 ","pages":"Article 109162"},"PeriodicalIF":4.8,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145518513","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}

引用次数: 0

Patient-specific prediction of 3D ablation zones via oncological feature-conditioned deep generative modeling: An in silico feasibility study 通过肿瘤特征条件下的深度生成建模对三维消融区域的患者特异性预测：一项计算机可行性研究。

IF 4.8 2区医学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computer methods and programs in biomedicine

Pub Date : 2026-02-01 Epub Date: 2025-12-01 DOI: 10.1016/j.cmpb.2025.109185

Hyo-Jin Kim , Truong Nhut Huynh , Ji-Won Lee , In-Seon Lee

Background and Objective:

Accurate prediction of the ablation zone is essential for effective radiofrequency ablation (RFA) therapy. However, current clinical guidelines often rely on oversimplified ellipsoidal assumptions based on homogeneous tissue properties, failing to reflect the patient-specific variability in cancer morphology. This study proposes a deep generative modeling framework conditioned on MRI-derived oncological features and electrode placement information to generate in silico 3D ablation zone predictions.

Methods:

Our model utilizes a U-Net-based generator within a conditional DCGAN framework to capture irregular, non-ellipsoidal thermal patterns arising from tissue-dependent thermal and electrical heterogeneity. Trained on synthetic ablation data generated from validated numerical simulations incorporating MRI-derived cancer morphology, it implicitly learns the influence of biophysical heterogeneity. The model accuracy was evaluated using intersection-over-union (IoU), precision, sensitivity, and specificity metrics and tested on unseen patient data.

Results:

The model achieved high accuracy with an IoU of 90.92

\pm

3.54% and generated predictions in under 0.5 s, suggesting potential for integration into clinical decision-support workflows. It also demonstrated strong generalization to unseen patient data.

Conclusion:

This study demonstrates the feasibility of deep generative modeling conditioned on MRI-derived oncological features for generating in silico 3D ablation zones. By leveraging biophysically validated simulation data, this model enables the virtual simulation and comparison of treatment outcomes from various electrode configurations and cancer morphologies, thereby refining the therapeutic strategy for each individual.

背景与目的：准确预测消融区是有效射频消融（RFA）治疗的关键。然而，目前的临床指南往往依赖于基于均匀组织特性的过于简化的椭球假设，未能反映癌症形态的患者特异性变异性。本研究提出了一个基于mri衍生的肿瘤特征和电极放置信息的深度生成建模框架，以生成硅三维消融区预测。方法：我们的模型在条件DCGAN框架内利用基于u - net的发生器来捕获由组织相关的热和电非均质性引起的不规则、非椭球形热模式。在经过验证的数值模拟生成的综合消融数据上进行训练，并结合mri衍生的癌症形态，它隐含地学习生物物理异质性的影响。使用交叉-超联合（IoU）、精度、敏感性和特异性指标评估模型的准确性，并对未见过的患者数据进行测试。结果：该模型获得了较高的准确率，IoU为90.92±3.54%，预测时间小于0.5 s，具有整合到临床决策支持工作流程中的潜力。它还显示了对未见过的患者数据的强大泛化。结论：本研究证明了基于mri衍生肿瘤特征的深度生成建模的可行性，该建模可用于生成硅三维消融区。通过利用生物物理验证的模拟数据，该模型可以对不同电极配置和癌症形态的治疗结果进行虚拟模拟和比较，从而细化每个个体的治疗策略。

{"title":"Patient-specific prediction of 3D ablation zones via oncological feature-conditioned deep generative modeling: An in silico feasibility study","authors":"Hyo-Jin Kim , Truong Nhut Huynh , Ji-Won Lee , In-Seon Lee","doi":"10.1016/j.cmpb.2025.109185","DOIUrl":"10.1016/j.cmpb.2025.109185","url":null,"abstract":"<div><h3>Background and Objective:</h3><div>Accurate prediction of the ablation zone is essential for effective radiofrequency ablation (RFA) therapy. However, current clinical guidelines often rely on oversimplified ellipsoidal assumptions based on homogeneous tissue properties, failing to reflect the patient-specific variability in cancer morphology. This study proposes a deep generative modeling framework conditioned on MRI-derived oncological features and electrode placement information to generate in silico 3D ablation zone predictions.</div></div><div><h3>Methods:</h3><div>Our model utilizes a U-Net-based generator within a conditional DCGAN framework to capture irregular, non-ellipsoidal thermal patterns arising from tissue-dependent thermal and electrical heterogeneity. Trained on synthetic ablation data generated from validated numerical simulations incorporating MRI-derived cancer morphology, it implicitly learns the influence of biophysical heterogeneity. The model accuracy was evaluated using intersection-over-union (IoU), precision, sensitivity, and specificity metrics and tested on unseen patient data.</div></div><div><h3>Results:</h3><div>The model achieved high accuracy with an IoU of 90.92<span><math><mo>±</mo></math></span>3.54% and generated predictions in under 0.5 s, suggesting potential for integration into clinical decision-support workflows. It also demonstrated strong generalization to unseen patient data.</div></div><div><h3>Conclusion:</h3><div>This study demonstrates the feasibility of deep generative modeling conditioned on MRI-derived oncological features for generating in silico 3D ablation zones. By leveraging biophysically validated simulation data, this model enables the virtual simulation and comparison of treatment outcomes from various electrode configurations and cancer morphologies, thereby refining the therapeutic strategy for each individual.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"275 ","pages":"Article 109185"},"PeriodicalIF":4.8,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145667469","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}

引用次数: 0

Fluid–structure interaction analysis of carotid artery stenosis: Impact of concentric versus eccentric morphologies on hemodynamic parameters 颈动脉狭窄的流固相互作用分析：同心与偏心形态对血流动力学参数的影响。

IF 4.8 2区医学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computer methods and programs in biomedicine

Pub Date : 2026-02-01 Epub Date: 2025-11-15 DOI: 10.1016/j.cmpb.2025.109168

Kshitij Shakya , P.M. Nabeel , Jayaraj Joseph , Shubhajit Roy Chowdhury

Background and Objective

: The research on arterial stenosis has gained rapid attention in the last decade due to its severe consequences and chronicity. The morphological analysis of a stenosis plays a crucial role in understanding the physiological characteristics and its treatment strategies because alteration of stenoses’ shapes and their structure significantly changes the hemodynamics in the nearby region and may affect their progression.

Methods

: In this study, three common shapes of stenosis — round, oval, and elongated — located at the vulnerable point, the carotid sinus, have been discussed, along with their shape concentricity and eccentricity. The hemodynamics of these shapes of stenosis, along with their concentricity and eccentricity, have been explored in this study. A total of six cases (three different shapes of stenosis, and their concentricity and eccentricity) were discussed, and their hemodynamic parameters were described. The study performed a numerical analysis under Fluid-Structure Interaction (FSI) multiphysics.

Results

: After comparing all the hemodynamic parameters, it was found that which parameter is most suitable for distinguishing between different stenosis shapes. To analyse this, a phase shift and Root Mean Square Deviation (RMSD) were calculated for every parameter. It was found that the phase shift in case of velocity profile showed a maximum value of 15.79% and 10.53% in concentric and eccentric round cases, respectively, distinguishing it from the other two shapes. Similarly, the phase shift profile in displacement separated out the elongated shape, which had the minimum values. The eccentric and concentric round, elongated, and oval shapes were clearly distinguished when they were analysed using RMSD velocity and displacement profile, as they had a unique value of RMSD in each case.

Conclusion

: By analysing different hemodynamic parameters in a stenosed carotid artery, this paper showed that a specific shape of stenosis can be distinguished from the others. This information is valuable for researchers and doctors dealing with carotid stenosis and can be a useful reference during preoperative planning and surgical intervention for stenosis.

背景与目的：动脉狭窄因其后果严重、慢性等特点，近十年来引起了人们的广泛关注。狭窄的形态学分析对了解其生理特征和治疗策略起着至关重要的作用，因为狭窄的形状和结构的改变会显著改变附近区域的血流动力学，并可能影响其进展。方法：在本研究中，讨论了三种常见的狭窄形状-圆形，椭圆形和细长型-位于易损点颈动脉窦，以及它们的形状同心性和偏心性。这些狭窄形状的血流动力学，以及它们的同心度和偏心度，已经在本研究中进行了探讨。本文讨论了6例（3种不同形状的狭窄及其同心度和偏心率），并对其血流动力学参数进行了描述。在流固耦合（FSI）多物理场下进行了数值分析。结果：通过比较各种血流动力学参数，找出最适合区分不同狭窄形态的参数。为了分析这一点，计算每个参数的相移和均方根偏差（RMSD）。结果表明，速度剖面相移在同心圆和偏心圆情况下的最大值分别为15.79%和10.53%，与其他两种形状有所区别。同样，位移相移曲线也分离出具有最小值的细长形状。当使用RMSD速度和位移剖面分析时，可以清楚地区分偏心和同心圆、细长和椭圆形，因为它们在每种情况下都具有独特的RMSD值。结论：通过分析颈动脉狭窄的不同血流动力学参数，可以区分出特定形状的狭窄。这些信息对于研究颈动脉狭窄的研究人员和医生来说是有价值的，可以作为术前计划和狭窄手术干预的有用参考。

{"title":"Fluid–structure interaction analysis of carotid artery stenosis: Impact of concentric versus eccentric morphologies on hemodynamic parameters","authors":"Kshitij Shakya , P.M. Nabeel , Jayaraj Joseph , Shubhajit Roy Chowdhury","doi":"10.1016/j.cmpb.2025.109168","DOIUrl":"10.1016/j.cmpb.2025.109168","url":null,"abstract":"<div><h3>Background and Objective</h3><div>: The research on arterial stenosis has gained rapid attention in the last decade due to its severe consequences and chronicity. The morphological analysis of a stenosis plays a crucial role in understanding the physiological characteristics and its treatment strategies because alteration of stenoses’ shapes and their structure significantly changes the hemodynamics in the nearby region and may affect their progression.</div></div><div><h3>Methods</h3><div>: In this study, three common shapes of stenosis — round, oval, and elongated — located at the vulnerable point, the carotid sinus, have been discussed, along with their shape concentricity and eccentricity. The hemodynamics of these shapes of stenosis, along with their concentricity and eccentricity, have been explored in this study. A total of six cases (three different shapes of stenosis, and their concentricity and eccentricity) were discussed, and their hemodynamic parameters were described. The study performed a numerical analysis under Fluid-Structure Interaction (FSI) multiphysics.</div></div><div><h3>Results</h3><div>: After comparing all the hemodynamic parameters, it was found that which parameter is most suitable for distinguishing between different stenosis shapes. To analyse this, a phase shift and Root Mean Square Deviation (RMSD) were calculated for every parameter. It was found that the phase shift in case of velocity profile showed a maximum value of 15.79% and 10.53% in concentric and eccentric round cases, respectively, distinguishing it from the other two shapes. Similarly, the phase shift profile in displacement separated out the elongated shape, which had the minimum values. The eccentric and concentric round, elongated, and oval shapes were clearly distinguished when they were analysed using RMSD velocity and displacement profile, as they had a unique value of RMSD in each case.</div></div><div><h3>Conclusion</h3><div>: By analysing different hemodynamic parameters in a stenosed carotid artery, this paper showed that a specific shape of stenosis can be distinguished from the others. This information is valuable for researchers and doctors dealing with carotid stenosis and can be a useful reference during preoperative planning and surgical intervention for stenosis.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"274 ","pages":"Article 109168"},"PeriodicalIF":4.8,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145548580","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}

引用次数: 0

The impact of arterial cannula direction in central venoarterial extracorporeal membrane oxygenation on aortic hemodynamic characteristics under various perfusion conditions 中心静脉-动脉体外膜氧合中动脉插管方向对不同灌注条件下主动脉血流动力学特性的影响。

IF 4.8 2区医学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computer methods and programs in biomedicine

Pub Date : 2026-02-01 Epub Date: 2025-11-11 DOI: 10.1016/j.cmpb.2025.109164

Yifeng Xi , Yuan Li , Hongyu Wang , Xiaofei Wang , Bingyang Ji , Zengsheng Chen

Objective

To systematically evaluate the influence of cannula direction in central venoarterial extracorporeal membrane oxygenation (CVA ECMO) on aortic hemodynamics and oxygen transport through a multiparametric comparative analysis under varying cardiac output conditions.

Methods

Computational fluid dynamics methods were applied to analyze five cannula directions (Down, Vertical, Inner, Center, Outer) ECMO models under four cardiac insufficiency conditions. Evaluation parameters included hyperoxic blood distribution, residence time, and scalar shear stress (SSS), oxygen saturation distribution, hemolysis risk (HI), and thrombosis risk.

Results

Both cardiac recovery and cannula direction have a significant effect on the internal environment of aortic blood flow. The recovery of cardiac function can lead to coronary artery hypoxia. The hypoxic blood from the heart and the hyperoxic blood from the ECMO will underwent mixing before reaching the descending aorta and then perfuse to the lower extremities. The cannula directed Down and Vertical improves coronary oxygen supply (oxygen saturation of 93%-98% on ECMO 90% support), but increased aortic wall shear stress (WSS), with mean WSS increase of 27%-54.7%. When cannulas were directed toward the Inner and Center, it significantly reduced the volume of the high SSS region of the vessel and the risk of hemolysis (86% reduction in the volume of the high SSS and 44%-46% reduction in the mean HI when the level of ECMO support was reduced from 80% to 60%) but had a greater adverse impact on coronary oxygen supply (coronary arteries were fully supplied by the heart). The oxygenation of the renal and iliac arteries was less affected by the direction of cannulation, but the risk of thrombosis was significantly increased at the abdominal aortic bifurcation.

Conclusion

The recovery of cardiac function initially results in coronary artery hypoxia, and as the cardiac ejection capacity recovered, it also leads to aortic arch branch vessel hypoxia. Adjusting the direction of cannulation can improve the supply of hyperoxic blood to the coronary arteries (cannula facing Down) or to the branches of the aortic arch (cannula facing Outer). This study provides an important hemodynamic basis for the optimization of ECMO cannulation strategy.

目的：通过多参数对比分析，系统评价不同心输出量条件下中央静脉体外膜氧合（CVA ECMO）插管方向对主动脉血流动力学和氧转运的影响。方法：应用计算流体力学方法分析4种心功能不全情况下5种导管方向（下、立、内、中、外）的ECMO模型。评价参数包括高氧血分布、停留时间、标量剪切应力（SSS）、血氧饱和度分布、溶血风险（HI）、血栓形成风险。结果：心脏恢复和导管方向对主动脉血流内环境均有显著影响。心功能恢复可导致冠状动脉缺氧。来自心脏的低氧血和来自ECMO的高氧血在到达降主动脉前进行混合，然后向下肢灌注。导管向下垂直可改善冠状动脉供氧（ECMO支持90%时血氧饱和度为93% ~ 98%），但增加了主动脉壁剪切应力（WSS），平均增加27% ~ 54.7%。当导管指向内层和中心时，它显著减少了血管高SSS区域的体积和溶血的风险（当ECMO支持水平从80%降低到60%时，高SSS区域的体积减少86%，平均HI减少44%-46%），但对冠状动脉供氧有更大的不利影响（冠状动脉由心脏完全供应）。置管方向对肾动脉和髂动脉氧合影响较小，但腹主动脉分叉处血栓形成风险明显增加。结论：心功能恢复最初导致冠状动脉缺氧，随着心脏射血能力的恢复，导致主动脉弓支血管缺氧。调整插管方向可改善冠状动脉（导管朝下）或主动脉弓分支（导管朝外）的高氧血供。本研究为优化ECMO插管策略提供了重要的血流动力学依据。

{"title":"The impact of arterial cannula direction in central venoarterial extracorporeal membrane oxygenation on aortic hemodynamic characteristics under various perfusion conditions","authors":"Yifeng Xi , Yuan Li , Hongyu Wang , Xiaofei Wang , Bingyang Ji , Zengsheng Chen","doi":"10.1016/j.cmpb.2025.109164","DOIUrl":"10.1016/j.cmpb.2025.109164","url":null,"abstract":"<div><h3>Objective</h3><div>To systematically evaluate the influence of cannula direction in central venoarterial extracorporeal membrane oxygenation (CVA ECMO) on aortic hemodynamics and oxygen transport through a multiparametric comparative analysis under varying cardiac output conditions.</div></div><div><h3>Methods</h3><div>Computational fluid dynamics methods were applied to analyze five cannula directions (Down, Vertical, Inner, Center, Outer) ECMO models under four cardiac insufficiency conditions. Evaluation parameters included hyperoxic blood distribution, residence time, and scalar shear stress (SSS), oxygen saturation distribution, hemolysis risk (HI), and thrombosis risk.</div></div><div><h3>Results</h3><div>Both cardiac recovery and cannula direction have a significant effect on the internal environment of aortic blood flow. The recovery of cardiac function can lead to coronary artery hypoxia. The hypoxic blood from the heart and the hyperoxic blood from the ECMO will underwent mixing before reaching the descending aorta and then perfuse to the lower extremities. The cannula directed Down and Vertical improves coronary oxygen supply (oxygen saturation of 93%-98% on ECMO 90% support), but increased aortic wall shear stress (WSS), with mean WSS increase of 27%-54.7%. When cannulas were directed toward the Inner and Center, it significantly reduced the volume of the high SSS region of the vessel and the risk of hemolysis (86% reduction in the volume of the high SSS and 44%-46% reduction in the mean HI when the level of ECMO support was reduced from 80% to 60%) but had a greater adverse impact on coronary oxygen supply (coronary arteries were fully supplied by the heart). The oxygenation of the renal and iliac arteries was less affected by the direction of cannulation, but the risk of thrombosis was significantly increased at the abdominal aortic bifurcation.</div></div><div><h3>Conclusion</h3><div>The recovery of cardiac function initially results in coronary artery hypoxia, and as the cardiac ejection capacity recovered, it also leads to aortic arch branch vessel hypoxia. Adjusting the direction of cannulation can improve the supply of hyperoxic blood to the coronary arteries (cannula facing Down) or to the branches of the aortic arch (cannula facing Outer). This study provides an important hemodynamic basis for the optimization of ECMO cannulation strategy.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"274 ","pages":"Article 109164"},"PeriodicalIF":4.8,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145548537","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}

引用次数: 0

A multi-constituent model of thrombosis for blood-contacting medical devices 血液接触医疗器械血栓形成的多组分模型。

IF 4.8 2区医学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computer methods and programs in biomedicine

Pub Date : 2026-02-01 Epub Date: 2025-11-19 DOI: 10.1016/j.cmpb.2025.109158

Yuning Lin , Yuzhou Cheng , Kaiyue Yang , Kun Luo , Jianren Fan , Ru Lin , Qiang Shu

Background and objectives:

Ventricular assist devices (VADs) are currently effective clinical interventions for treating diseases related to end-stage heart failure (HF), yet hemocompatibility-related complications persist as prevalent challenges. Among them, device-induced thrombosis may lead to severe consequences such as stroke, neurological events, pump replacement, and even mortality, making it critically important to predict thrombosis accurately. The primary objective of this study is to develop a thrombosis model capable of simulating thrombus formation within VADs.

Methods:

The proposed model integrates hemodynamics, platelet activity, and the coagulation cascade, wherein the cascade products regulate platelet activation, aggregation, and stabilization. To enable simulations at the device scale while preserving essential physiological mechanisms, a reduced-order coagulation cascade model was adopted. Furthermore, the model incorporates hemodynamic-thrombus interaction while thrombus breakdown due to the shear stress clearance is under consideration.

Results:

The model was first validated in a backward-facing step (BFS) geometry to assess its applicability under separated flow conditions. In terms of volumetric evolution, the simulation followed a trend consistent with experimental data, while the thrombus length and height matched the experimental measurements closely. The model was then applied to a left ventricular assist device (VAD) to explore thrombus formation mechanisms. Simulations at different flow rates revealed consistent thrombosis on the straightener blades, while initiation sites and growth dynamics were governed by local hemodynamics, platelet activation, and stabilization.

Conclusions:

The thrombosis model developed in this study enables the investigation of thrombus formation mechanisms and the identification of potential high-risk regions within VADs under varying flow conditions. It provides a basis for future experimental validation and has potential utility for optimizing VAD design and informing patient-specific risk assessment.

背景和目的：心室辅助装置（VADs）是目前治疗终末期心力衰竭（HF）相关疾病的有效临床干预措施，但血液相容性相关并发症仍然是普遍存在的挑战。其中，器械诱发的血栓形成可能导致中风、神经系统事件、泵置换甚至死亡等严重后果，因此准确预测血栓形成至关重要。本研究的主要目的是建立一个能够模拟VADs内血栓形成的血栓形成模型。方法：该模型整合了血流动力学、血小板活性和凝血级联，其中级联产物调节血小板的活化、聚集和稳定。为了能够在设备规模上进行模拟，同时保留基本的生理机制，采用了降阶凝血级联模型。此外，该模型考虑了血流动力学-血栓相互作用，同时考虑了剪切应力清除引起的血栓破裂。结果：该模型首先在后向台阶（BFS）几何结构中进行了验证，以评估其在分离流动条件下的适用性。在体积演化方面，模拟的趋势与实验数据一致，血栓的长度和高度与实验测量值非常吻合。然后将该模型应用于左心室辅助装置（VAD）以探索血栓形成机制。不同流速下的模拟结果表明，矫直机叶片上的血栓形成是一致的，而起始位置和生长动力学受局部血流动力学、血小板活化和稳定的控制。结论：本研究建立的血栓形成模型可以研究不同血流条件下VADs内血栓形成机制和潜在高危区域的识别。它为未来的实验验证提供了基础，并具有优化VAD设计和告知患者特定风险评估的潜在效用。

{"title":"A multi-constituent model of thrombosis for blood-contacting medical devices","authors":"Yuning Lin , Yuzhou Cheng , Kaiyue Yang , Kun Luo , Jianren Fan , Ru Lin , Qiang Shu","doi":"10.1016/j.cmpb.2025.109158","DOIUrl":"10.1016/j.cmpb.2025.109158","url":null,"abstract":"<div><h3>Background and objectives:</h3><div>Ventricular assist devices (VADs) are currently effective clinical interventions for treating diseases related to end-stage heart failure (HF), yet hemocompatibility-related complications persist as prevalent challenges. Among them, device-induced thrombosis may lead to severe consequences such as stroke, neurological events, pump replacement, and even mortality, making it critically important to predict thrombosis accurately. The primary objective of this study is to develop a thrombosis model capable of simulating thrombus formation within VADs.</div></div><div><h3>Methods:</h3><div>The proposed model integrates hemodynamics, platelet activity, and the coagulation cascade, wherein the cascade products regulate platelet activation, aggregation, and stabilization. To enable simulations at the device scale while preserving essential physiological mechanisms, a reduced-order coagulation cascade model was adopted. Furthermore, the model incorporates hemodynamic-thrombus interaction while thrombus breakdown due to the shear stress clearance is under consideration.</div></div><div><h3>Results:</h3><div>The model was first validated in a backward-facing step (BFS) geometry to assess its applicability under separated flow conditions. In terms of volumetric evolution, the simulation followed a trend consistent with experimental data, while the thrombus length and height matched the experimental measurements closely. The model was then applied to a left ventricular assist device (VAD) to explore thrombus formation mechanisms. Simulations at different flow rates revealed consistent thrombosis on the straightener blades, while initiation sites and growth dynamics were governed by local hemodynamics, platelet activation, and stabilization.</div></div><div><h3>Conclusions:</h3><div>The thrombosis model developed in this study enables the investigation of thrombus formation mechanisms and the identification of potential high-risk regions within VADs under varying flow conditions. It provides a basis for future experimental validation and has potential utility for optimizing VAD design and informing patient-specific risk assessment.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"274 ","pages":"Article 109158"},"PeriodicalIF":4.8,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145602764","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}

引用次数: 0

Novel machine learning framework for multidimensional biological age estimation reveals heterogeneous aging of organ systems 多维生物年龄估计的新机器学习框架揭示了器官系统的异质老化

IF 4.8 2区医学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computer methods and programs in biomedicine

Pub Date : 2026-02-01 Epub Date: 2025-11-20 DOI: 10.1016/j.cmpb.2025.109176

Qi Yu , Lijuan Da , Qian Ma , Yushu Huang , Yue Dong , Yuan Liu , Xiaoyu Li , Xifeng Wu , Zilin Li , Wenyuan Li

Objective

Existing biological age (BA) models often oversimplify aging’s complexity, offering single-dimensional metrics. However, these fail to capture the critical heterogeneity of aging across organs.. This study aims to develop a machine learning-based unified framework to assess and interpret multi-organ biological aging comprehensively.

Method

Using data from UK Biobank participants, we trained and integrated organ-specific BA estimates to assess multidimensional BA within an ensemble learning framework, and uncover distinct aging patterns.

Result

Our Fusion BA (an overall estimation of BA) was significantly correlated with chronological age (CA) (mean absolute error (MAE): 4.473 years; Pearson correlation: 0.718, P < 0.01). Accelerated Fusion BA derived from the ensemble model (contrast between Fusion BA and CA) predicted 10-year mortality (HR=1.504, 95 % CI: 1.438–1.574). Organ-specific BA correlated with organ disease risk and effectively captured distinct aging patterns.

Conclusion

This framework enables systemic and organ-specific aging assessment, provides actionable tools and insights for clinical risk.

现有的生物年龄（BA）模型往往过于简化了衰老的复杂性，提供了单一的指标。然而，这些未能捕捉到各器官衰老的关键异质性。本研究旨在建立一个基于机器学习的统一框架来全面评估和解释多器官生物衰老。方法使用来自UK Biobank参与者的数据，我们训练并整合了器官特异性BA估计，以在集成学习框架内评估多维BA，并揭示不同的衰老模式。结果tour Fusion BA （BA的总估计值）与实足年龄（CA）显著相关(平均绝对误差（MAE）： 4.473岁；Pearson相关性：0.718,P < 0.01)。来自集合模型的加速融合BA（融合BA和CA的对比）预测10年死亡率（HR=1.504, 95% CI: 1.438-1.574）。器官特异性BA与器官疾病风险相关，并有效捕获不同的衰老模式。结论该框架可实现系统性和器官特异性衰老评估，为临床风险提供可操作的工具和见解。

{"title":"Novel machine learning framework for multidimensional biological age estimation reveals heterogeneous aging of organ systems","authors":"Qi Yu , Lijuan Da , Qian Ma , Yushu Huang , Yue Dong , Yuan Liu , Xiaoyu Li , Xifeng Wu , Zilin Li , Wenyuan Li","doi":"10.1016/j.cmpb.2025.109176","DOIUrl":"10.1016/j.cmpb.2025.109176","url":null,"abstract":"<div><h3>Objective</h3><div>Existing biological age (BA) models often oversimplify aging’s complexity, offering single-dimensional metrics. However, these fail to capture the critical heterogeneity of aging across organs.. This study aims to develop a machine learning-based unified framework to assess and interpret multi-organ biological aging comprehensively.</div></div><div><h3>Method</h3><div>Using data from UK Biobank participants, we trained and integrated organ-specific BA estimates to assess multidimensional BA within an ensemble learning framework, and uncover distinct aging patterns.</div></div><div><h3>Result</h3><div>Our Fusion BA (an overall estimation of BA) was significantly correlated with chronological age (CA) (mean absolute error (MAE): 4.473 years; Pearson correlation: 0.718, <em>P</em> < 0.01). Accelerated Fusion BA derived from the ensemble model (contrast between Fusion BA and CA) predicted 10-year mortality (HR=1.504, 95 % CI: 1.438–1.574). Organ-specific BA correlated with organ disease risk and effectively captured distinct aging patterns.</div></div><div><h3>Conclusion</h3><div>This framework enables systemic and organ-specific aging assessment, provides actionable tools and insights for clinical risk.</div></div>","PeriodicalId":10624,"journal":{"name":"Computer methods and programs in biomedicine","volume":"275 ","pages":"Article 109176"},"PeriodicalIF":4.8,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145682465","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}

引用次数: 0