Journal of Pathology Informatics最新文献

英文中文

A machine learning model of lamina propria fibrosis in eosinophilic esophagitis for prediction of fibrostenotic disease 嗜酸性粒细胞性食管炎固有层纤维化的机器学习模型预测纤维狭窄性疾病

Q2 Medicine

Journal of Pathology Informatics

Pub Date : 2026-01-01 Epub Date: 2025-12-22 DOI: 10.1016/j.jpi.2025.100538

Priyadharshini Sivasubramaniam , Abdelrahman Shabaan , Rofyda Elhalaby , Bashar Hasan , Ameya A. Patil , Saadiya Nazli , Adilson DaCosta , Byoung Uk Park , Lindsey Smith , Taofic Mounajjed , Stephen M. Lagana , Chamil Codipilly , Puanani Hopson , Imad Absah , Christopher P. Hartley , Rondell P. Graham , Roger K. Moreira

Background

Eosinophilic esophagitis (EoE) is a chronic immune-mediated disease that can progress to fibrostenotic complications. Lamina propria fibrosis (LPF) plays a critical role in this progression but is difficult to assess reliably in routine biopsies. We aimed to develop and validate an artificial intelligence (AI) model to quantify LPF on hematoxylin and eosin (H&E)-stained slides and to evaluate its ability to predict fibrostenotic disease.

Methods

We used a cloud-based platform (Aiforia Inc., Cambridge, MA, USA) to train a supervised AI model to recognize several histological features of EoE, including LPF. Our validation cohort consisted of 213 esophageal biopsy whole-slide images, including 100 adult and 113 pediatric samples with mucosal eosinophilia, which were prospectively evaluated in our anatomic pathology service between 2020 and 2021 using a standardized histological scoring system. AI-based LPF scores were correlated with the development of fibrostenotic disease on subsequent endoscopies after a median follow-up time of 31.4 months.

Results

The AI fibrosis score correlated with pathologist-determined LPF (Spearman's Rs = 0.64–0.69, p < 0.0001) and outperformed pathologists' assessments in predicting fibrostenotic outcomes. Higher AI fibrosis scores were associated with the development of rings, strictures, and need for dilatation on follow-up (p < 0.01), including in cases deemed histologically inadequate by pathologists and in the subgroup without prior strictures. In a Cox Proportional-Hazards model, the AI fibrosis score was an independent predictor of strictures (C-index = 0.73, p = 0.004). Importantly, meaningful predictions were achievable with smaller amounts of lamina propria than traditionally deemed sufficient.

Conclusion

This study demonstrates that AI-based quantification of LPF on routine H&E slides provides an objective and clinically meaningful assessment of fibrosis in EoE. The AI fibrosis score predicts fibrostenotic disease more consistently than conventional pathology evaluation and may improve risk stratification even in limited biopsy samples. Integration of digital pathology tools may enhance histological assessment of fibrosis in EoE and support clinical decision-making.

嗜酸性粒细胞性食管炎（EoE）是一种慢性免疫介导的疾病，可发展为纤维狭窄并发症。固有层纤维化（LPF）在这种进展中起关键作用，但在常规活检中难以可靠地评估。我们旨在开发和验证人工智能（AI）模型，以量化苏木精和伊红（H&；E）染色玻片上的LPF，并评估其预测纤维狭窄性疾病的能力。方法我们使用基于云的平台（Aiforia Inc., Cambridge， MA， USA）来训练一个有监督的AI模型来识别EoE的几种组织学特征，包括LPF。我们的验证队列包括213张食管活检全切片图像，包括100例成人和113例儿童粘膜嗜酸性粒细胞增生样本，这些样本在2020年至2021年期间在我们的解剖病理学服务中使用标准化组织学评分系统进行前瞻性评估。在中位随访31.4 个月后，基于人工智能的LPF评分与随后的内窥镜检查中纤维狭窄性疾病的发展相关。结果AI纤维化评分与病理学确定的LPF相关（Spearman’s Rs = 0.64-0.69,p <； 0.0001），在预测纤维狭窄结局方面优于病理学评估。较高的AI纤维化评分与环、狭窄的发展和随访时需要扩张相关（p <； 0.01），包括病理学家认为组织学不充分的病例和先前没有狭窄的亚组。在Cox比例风险模型中，AI纤维化评分是狭窄的独立预测因子（c指数 = 0.73,p = 0.004）。重要的是，有意义的预测是可以实现较少的固有层比传统认为足够的。结论本研究表明，基于人工智能的常规H&；E载玻片LPF定量为评估EoE纤维化提供了客观且具有临床意义的方法。AI纤维化评分预测纤维狭窄性疾病比常规病理评估更一致，即使在有限的活检样本中也可能改善风险分层。数字病理工具的整合可以增强EoE纤维化的组织学评估并支持临床决策。

{"title":"A machine learning model of lamina propria fibrosis in eosinophilic esophagitis for prediction of fibrostenotic disease","authors":"Priyadharshini Sivasubramaniam , Abdelrahman Shabaan , Rofyda Elhalaby , Bashar Hasan , Ameya A. Patil , Saadiya Nazli , Adilson DaCosta , Byoung Uk Park , Lindsey Smith , Taofic Mounajjed , Stephen M. Lagana , Chamil Codipilly , Puanani Hopson , Imad Absah , Christopher P. Hartley , Rondell P. Graham , Roger K. Moreira","doi":"10.1016/j.jpi.2025.100538","DOIUrl":"10.1016/j.jpi.2025.100538","url":null,"abstract":"<div><h3>Background</h3><div>Eosinophilic esophagitis (EoE) is a chronic immune-mediated disease that can progress to fibrostenotic complications. Lamina propria fibrosis (LPF) plays a critical role in this progression but is difficult to assess reliably in routine biopsies. We aimed to develop and validate an artificial intelligence (AI) model to quantify LPF on hematoxylin and eosin (H&E)-stained slides and to evaluate its ability to predict fibrostenotic disease.</div></div><div><h3>Methods</h3><div>We used a cloud-based platform (Aiforia Inc., Cambridge, MA, USA) to train a supervised AI model to recognize several histological features of EoE, including LPF. Our validation cohort consisted of 213 esophageal biopsy whole-slide images, including 100 adult and 113 pediatric samples with mucosal eosinophilia, which were prospectively evaluated in our anatomic pathology service between 2020 and 2021 using a standardized histological scoring system. AI-based LPF scores were correlated with the development of fibrostenotic disease on subsequent endoscopies after a median follow-up time of 31.4 months.</div></div><div><h3>Results</h3><div>The AI fibrosis score correlated with pathologist-determined LPF (Spearman's <em>R</em>s = 0.64–0.69, <em>p</em> < 0.0001) and outperformed pathologists' assessments in predicting fibrostenotic outcomes. Higher AI fibrosis scores were associated with the development of rings, strictures, and need for dilatation on follow-up (<em>p</em> < 0.01), including in cases deemed histologically inadequate by pathologists and in the subgroup without prior strictures. In a Cox Proportional-Hazards model, the AI fibrosis score was an independent predictor of strictures (C-index = 0.73, <em>p</em> = 0.004). Importantly, meaningful predictions were achievable with smaller amounts of lamina propria than traditionally deemed sufficient.</div></div><div><h3>Conclusion</h3><div>This study demonstrates that AI-based quantification of LPF on routine H&E slides provides an objective and clinically meaningful assessment of fibrosis in EoE. The AI fibrosis score predicts fibrostenotic disease more consistently than conventional pathology evaluation and may improve risk stratification even in limited biopsy samples. Integration of digital pathology tools may enhance histological assessment of fibrosis in EoE and support clinical decision-making.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"20 ","pages":"Article 100538"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977195","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Biological feature-based machine learning in histopathological images: a systematic review 组织病理学图像中基于生物特征的机器学习：系统综述

Q2 Medicine

Journal of Pathology Informatics

Pub Date : 2026-01-01 Epub Date: 2026-01-02 DOI: 10.1016/j.jpi.2025.100539

Stéphane Treillard , Robin Schwob , Sandrine Mouysset , Pierre Brousset , Sylvain Cussat-Blanc , Camille Franchet

Digital pathology has recently led to significant advancements in the field of microscopic image analysis, particularly regarding the increasing use of Deep Learning methods. These models represent the state-of-the-art in histopathological slide analysis, but Deep Learning features remain difficult to interpret, despite recent developments in post hoc explainability frameworks. In contrast, features extracted from biological objects—such as nuclei, cells or tissues—are supposed to be more grounded in pathologists' a priori knowledge. Accordingly, Machine Learning based on handcrafted features represents another paradigm of explainability and may stand as a complementary method to Deep Learning to assist pathologists.

In order to perceive how biological features have been used in hematoxylin & eosin microscopic images to address medical questions, we conducted a systematic review of articles published from January 2005 to May 2025, adhering to PRISMA guidelines.

A total of 97 articles were analyzed from the PubMed, IEEE, and ACM databases. Three primary categories of features—texture/color, morphology and topology—were both identified and thoroughly described. These features were most frequently derived from segmented cells and tissues in 80 and 28 studies, respectively. They were used to address seven types of medical questions: “normal vs diseased”, disease subtyping, tumor grading, phenotyping, object detection, prognosis and treatment-response prediction.

We discussed methodological and reporting limitations of these studies, highlighting the difficulty to assess the potential impact of such methods. Among the most common concerns, we found features difficult to interpret, data leakage, and inadequate sample sizes. Nevertheless, we also focused on promising domain-inspired feature engineering that provides better explainability and specificity. This kind of features associated with more methodological rigor may increase the relevance and reliability of AI models, and also raise new research avenues in pathology.

数字病理学最近在显微图像分析领域取得了重大进展，特别是关于越来越多地使用深度学习方法。这些模型代表了组织病理学切片分析的最新技术，但深度学习的特征仍然难以解释，尽管最近在事后解释框架方面取得了进展。相比之下，从生物物体中提取的特征——如细胞核、细胞或组织——被认为是基于病理学家的先验知识。因此，基于手工特征的机器学习代表了另一种可解释性范式，可以作为深度学习的补充方法来帮助病理学家。为了了解如何利用苏木精伊红显微图像中的生物学特征来解决医学问题，我们根据PRISMA指南对2005年1月至2025年5月发表的文章进行了系统回顾。共分析了来自PubMed、IEEE和ACM数据库的97篇文章。三个主要类别的特征-纹理/颜色，形态和拓扑-都被识别和彻底描述。在80项和28项研究中，这些特征最常来自分节的细胞和组织。他们被用来解决七种类型的医学问题：“正常与患病”、疾病亚型、肿瘤分级、表型、目标检测、预后和治疗反应预测。我们讨论了这些研究的方法和报告局限性，强调了评估这些方法的潜在影响的困难。在最常见的问题中，我们发现了难以解释的特性、数据泄漏和样本量不足。尽管如此，我们也专注于有前途的领域启发特征工程，它提供了更好的可解释性和特异性。这种与更严谨的方法相关的特征可能会增加人工智能模型的相关性和可靠性，并为病理学提供新的研究途径。

{"title":"Biological feature-based machine learning in histopathological images: a systematic review","authors":"Stéphane Treillard , Robin Schwob , Sandrine Mouysset , Pierre Brousset , Sylvain Cussat-Blanc , Camille Franchet","doi":"10.1016/j.jpi.2025.100539","DOIUrl":"10.1016/j.jpi.2025.100539","url":null,"abstract":"<div><div>Digital pathology has recently led to significant advancements in the field of microscopic image analysis, particularly regarding the increasing use of Deep Learning methods. These models represent the state-of-the-art in histopathological slide analysis, but Deep Learning features remain difficult to interpret, despite recent developments in post hoc explainability frameworks. In contrast, features extracted from biological objects—such as nuclei, cells or tissues—are supposed to be more grounded in pathologists' a priori knowledge. Accordingly, Machine Learning based on handcrafted features represents another paradigm of explainability and may stand as a complementary method to Deep Learning to assist pathologists.</div><div>In order to perceive how biological features have been used in hematoxylin & eosin microscopic images to address medical questions, we conducted a systematic review of articles published from January 2005 to May 2025, adhering to PRISMA guidelines.</div><div>A total of 97 articles were analyzed from the PubMed, IEEE, and ACM databases. Three primary categories of features—texture/color, morphology and topology—were both identified and thoroughly described. These features were most frequently derived from segmented cells and tissues in 80 and 28 studies, respectively. They were used to address seven types of medical questions: “normal vs diseased”, disease subtyping, tumor grading, phenotyping, object detection, prognosis and treatment-response prediction.</div><div>We discussed methodological and reporting limitations of these studies, highlighting the difficulty to assess the potential impact of such methods. Among the most common concerns, we found features difficult to interpret, data leakage, and inadequate sample sizes. Nevertheless, we also focused on promising domain-inspired feature engineering that provides better explainability and specificity. This kind of features associated with more methodological rigor may increase the relevance and reliability of AI models, and also raise new research avenues in pathology.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"20 ","pages":"Article 100539"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146037907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Transforming multi-omics data into images for disease classification: A review of techniques and tools 将多组学数据转化为疾病分类的图像：技术和工具的综述

Q2 Medicine

Journal of Pathology Informatics

Pub Date : 2026-01-01 Epub Date: 2026-01-16 DOI: 10.1016/j.jpi.2026.100543

Ali Alyatimi , Muhammad Atif Iqbal , Vera Chung , Seid Miad Zandavi , Ali Anaissi

The integration of multi-omics data has become crucial in understanding the complexity of biological systems and disease mechanisms. However, the high dimensionality and heterogeneity of such data present significant analytical challenges. This review investigates the emerging approach of transforming multi-omics non-image data into image formats to facilitate the application of advanced deep learning techniques for disease classification and biomarker discovery. This article presents a scoping review of studies published between 2013 and 2024, focusing on techniques that convert multi-omics data into images. Various transformation methods, including t-SNE, kernel PCA, UMAP, FFT, and treemaps, were examined alongside deep learning models such as convolutional neural networks, autoencoders, support vector machines, graph convolutional networks, and graph neural networks. The transformation of omics data into image formats enables effective feature extraction and classification, with reported accuracies ranging from 75% to 99% across various studies. CNN-based models, in particular, demonstrated superior performance in integrating complex molecular interactions. Despite these advances, challenges such as overfitting, limited generalizability, and interpretability persist, especially given the diversity and complexity of multi-omics datasets. Finally, the transforming multi-omics data into images represents a promising direction in biomedical research, facilitating more profound insights into disease mechanisms and improving predictive modeling. Addressing current limitations through improved model interpretability, robust transformation methods, and larger, more diverse datasets will be essential for realizing the full potential of this approach in precision medicine.

多组学数据的整合对于理解生物系统和疾病机制的复杂性至关重要。然而，这些数据的高维度和异质性提出了重大的分析挑战。本文综述了将多组学非图像数据转换为图像格式的新兴方法，以促进先进的深度学习技术在疾病分类和生物标志物发现中的应用。本文介绍了2013年至2024年间发表的研究范围综述，重点是将多组学数据转换为图像的技术。各种变换方法，包括t-SNE、核PCA、UMAP、FFT和树图，与卷积神经网络、自动编码器、支持向量机、图卷积网络和图神经网络等深度学习模型一起进行了研究。将组学数据转换为图像格式可以实现有效的特征提取和分类，在各种研究中报道的准确率从75%到99%不等。特别是基于cnn的模型，在整合复杂分子相互作用方面表现出了卓越的性能。尽管取得了这些进展，但过度拟合、有限的通用性和可解释性等挑战仍然存在，特别是考虑到多组学数据集的多样性和复杂性。最后，将多组学数据转化为图像是生物医学研究的一个有前途的方向，有助于更深入地了解疾病机制并改进预测建模。通过改进的模型可解释性、稳健的转换方法和更大、更多样化的数据集来解决当前的局限性，对于实现这种方法在精准医学中的全部潜力至关重要。

{"title":"Transforming multi-omics data into images for disease classification: A review of techniques and tools","authors":"Ali Alyatimi , Muhammad Atif Iqbal , Vera Chung , Seid Miad Zandavi , Ali Anaissi","doi":"10.1016/j.jpi.2026.100543","DOIUrl":"10.1016/j.jpi.2026.100543","url":null,"abstract":"<div><div>The integration of multi-omics data has become crucial in understanding the complexity of biological systems and disease mechanisms. However, the high dimensionality and heterogeneity of such data present significant analytical challenges. This review investigates the emerging approach of transforming multi-omics non-image data into image formats to facilitate the application of advanced deep learning techniques for disease classification and biomarker discovery. This article presents a scoping review of studies published between 2013 and 2024, focusing on techniques that convert multi-omics data into images. Various transformation methods, including t-SNE, kernel PCA, UMAP, FFT, and treemaps, were examined alongside deep learning models such as convolutional neural networks, autoencoders, support vector machines, graph convolutional networks, and graph neural networks. The transformation of omics data into image formats enables effective feature extraction and classification, with reported accuracies ranging from 75% to 99% across various studies. CNN-based models, in particular, demonstrated superior performance in integrating complex molecular interactions. Despite these advances, challenges such as overfitting, limited generalizability, and interpretability persist, especially given the diversity and complexity of multi-omics datasets. Finally, the transforming multi-omics data into images represents a promising direction in biomedical research, facilitating more profound insights into disease mechanisms and improving predictive modeling. Addressing current limitations through improved model interpretability, robust transformation methods, and larger, more diverse datasets will be essential for realizing the full potential of this approach in precision medicine.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"20 ","pages":"Article 100543"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146188859","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Developing a smart and scalable tool for histopathological education—PATe 2.0 开发一个智能和可扩展的组织病理学教育工具- pate 2.0

Q2 Medicine

Journal of Pathology Informatics

Pub Date : 2026-01-01 Epub Date: 2025-12-05 DOI: 10.1016/j.jpi.2025.100535

Lina Winter , Annalena Artinger , Hendrik Böck , Vignesh Ramakrishnan , Bruno Reible , Jan Albin , Peter J. Schüffler , Georgios Raptis , Christoph Brochhausen

Digital microscopy plays a crucial role in pathology education, providing scalable and standardized access to learning resources. In response, we present PATe 2.0, a scalable redeveloped web-application of the former PATe system from 2015. PATe 2.0 was developed using an agile, iterative process and built on a microservices architecture to ensure modularity, scalability, and reliability. It integrates a modern web-based user interface optimized for desktop and tablet use and automates key workflows such as whole-slide image uploads and processing. Performance tests demonstrated that PATe 2.0 significantly reduces tile request times compared to PATe, despite handling larger tiles. The platform supports open formats like DICOM and OpenSlide, enhancing its interoperability and adaptability across institutions. PATe 2.0 represents a robust digital microscopy solution in pathology education enhancing usability, performance, and flexibility. Its design enables future integration of research algorithms and highlights it as a pivotal tool for advancing pathology education and research.

数字显微镜在病理学教育中起着至关重要的作用，提供了可扩展和标准化的学习资源。作为回应，我们提出了PATe 2.0，这是2015年以前的PATe系统的可扩展的重新开发的web应用程序。PATe 2.0是使用敏捷的迭代过程开发的，并构建在微服务体系结构上，以确保模块化、可伸缩性和可靠性。它集成了一个现代的基于web的用户界面，为桌面和平板电脑的使用进行了优化，并自动化了关键的工作流程，如整张幻灯片图像的上传和处理。性能测试表明，尽管处理的贴图更大，但与PATe相比，PATe 2.0显著减少了贴图请求时间。该平台支持DICOM和OpenSlide等开放格式，增强了其跨机构的互操作性和适应性。PATe 2.0代表了病理学教育中强大的数字显微镜解决方案，增强了可用性，性能和灵活性。它的设计使未来的研究算法的整合，并突出了它作为一个关键的工具，推进病理教育和研究。

{"title":"Developing a smart and scalable tool for histopathological education—PATe 2.0","authors":"Lina Winter , Annalena Artinger , Hendrik Böck , Vignesh Ramakrishnan , Bruno Reible , Jan Albin , Peter J. Schüffler , Georgios Raptis , Christoph Brochhausen","doi":"10.1016/j.jpi.2025.100535","DOIUrl":"10.1016/j.jpi.2025.100535","url":null,"abstract":"<div><div>Digital microscopy plays a crucial role in pathology education, providing scalable and standardized access to learning resources. In response, we present PATe 2.0, a scalable redeveloped web-application of the former PATe system from 2015. PATe 2.0 was developed using an agile, iterative process and built on a microservices architecture to ensure modularity, scalability, and reliability. It integrates a modern web-based user interface optimized for desktop and tablet use and automates key workflows such as whole-slide image uploads and processing. Performance tests demonstrated that PATe 2.0 significantly reduces tile request times compared to PATe, despite handling larger tiles. The platform supports open formats like DICOM and OpenSlide, enhancing its interoperability and adaptability across institutions. PATe 2.0 represents a robust digital microscopy solution in pathology education enhancing usability, performance, and flexibility. Its design enables future integration of research algorithms and highlights it as a pivotal tool for advancing pathology education and research.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"20 ","pages":"Article 100535"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791667","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Evaluating the robustness of slide-level AI predictions on out-of-focus whole slide images: A retrospective observational study 评估幻灯片级AI对失焦整张幻灯片图像预测的稳健性：一项回顾性观察研究

Q2 Medicine

Journal of Pathology Informatics

Pub Date : 2025-11-01 Epub Date: 2025-09-16 DOI: 10.1016/j.jpi.2025.100518

Ho Heon Kim , Young Sin Ko , Won Chan Jeong , Seokju Yun , Kyungeun Kim

Background

Blurriness in whole slide images (WSIs) is a common issue in digital pathology. Whereas severe blurriness is known to degrade artificial intelligence (AI) model performance, the impact of typical levels of blurriness observed in real-world settings remains unclear.

Objectives

To evaluate the effect of WSI blurring on robustness of AI predictions in real-world settings.

Methods

A retrospective study was conducted using 7529 WSIs and the corresponding AI predictions from 4 AI models trained on data from 2 scanners and 2 organs. The WSIs were categorized into concordant and discordant groups based on the AI prediction accuracy. Analyses included: (1) comparing blur metrics between groups, (2) determining the odds ratio between the proportions of blurry patch in WSIs and prediction concordance, (3) assessing model performance across various blur intensities, and (4) examining the similarity of slide- and patch-level embeddings across focal planes using Z-stacks.

Results

Regarding each organ–scanner pair, the average wavelet score and Laplacian variance did not show statistically significant differences between the two groups and no significant association was observed between prediction concordance and the proportion of blurry regions (p > 0.05, except one pair). Model performance remained robust even at a high blur level (radius = 1), where the patch image had a Laplacian variance of 133.14 and a wavelet score of 1667.98, corresponding to the top 8.6% and 12.15% of blurriness, respectively, in our dataset. In addition, embedding analysis across focal planes using Z-stacks revealed that both patch- and slide-level representations were preserved up to ±3 μm. Slide-level embeddings consistently exhibited cosine similarity values above 0.99.

Conclusions

These findings empirically suggest that the typical levels of WSI blurriness encountered in clinical practice may not significantly compromise the robustness of slide-level AI classification.

背景模糊在整个幻灯片图像（wsi）是一个常见的问题，在数字病理学。虽然已知严重的模糊会降低人工智能（AI）模型的性能，但在现实环境中观察到的典型模糊水平的影响尚不清楚。目的评估WSI模糊对现实世界中人工智能预测鲁棒性的影响。方法利用2台扫描仪和2个器官数据训练的4个人工智能模型的7529个wsi和相应的人工智能预测进行回顾性研究。根据人工智能预测精度将wsi分为一致性组和不一致性组。分析包括：(1)比较各组之间的模糊度量，(2)确定wsi中模糊斑块比例与预测一致性之间的比值比，(3)评估不同模糊强度下的模型性能，以及(4)使用z堆栈检查滑动和斑块级嵌入在焦平面上的相似性。结果各脏器扫描对的平均小波评分和拉普拉斯方差在两组间差异无统计学意义，预测一致性与模糊区域比例无显著相关性（p >； 0.05，除一对外）。即使在高模糊水平（半径 = 1）下，模型性能仍然保持稳健，其中patch图像的拉普拉斯方差为133.14，小波评分为1667.98，分别对应于我们数据集中前8.6%和12.15%的模糊程度。此外，使用z堆叠进行的跨焦平面嵌入分析显示，贴片和幻灯片级表示在±3 μm范围内都保持不变。幻灯片级嵌入的余弦相似度值始终高于0.99。这些研究结果表明，临床实践中遇到的典型WSI模糊程度可能不会显著影响幻灯片级AI分类的稳健性。

{"title":"Evaluating the robustness of slide-level AI predictions on out-of-focus whole slide images: A retrospective observational study","authors":"Ho Heon Kim , Young Sin Ko , Won Chan Jeong , Seokju Yun , Kyungeun Kim","doi":"10.1016/j.jpi.2025.100518","DOIUrl":"10.1016/j.jpi.2025.100518","url":null,"abstract":"<div><h3>Background</h3><div>Blurriness in whole slide images (WSIs) is a common issue in digital pathology. Whereas severe blurriness is known to degrade artificial intelligence (AI) model performance, the impact of typical levels of blurriness observed in real-world settings remains unclear.</div></div><div><h3>Objectives</h3><div>To evaluate the effect of WSI blurring on robustness of AI predictions in real-world settings.</div></div><div><h3>Methods</h3><div>A retrospective study was conducted using 7529 WSIs and the corresponding AI predictions from 4 AI models trained on data from 2 scanners and 2 organs. The WSIs were categorized into concordant and discordant groups based on the AI prediction accuracy. Analyses included: (1) comparing blur metrics between groups, (2) determining the odds ratio between the proportions of blurry patch in WSIs and prediction concordance, (3) assessing model performance across various blur intensities, and (4) examining the similarity of slide- and patch-level embeddings across focal planes using <em>Z</em>-stacks.</div></div><div><h3>Results</h3><div>Regarding each organ–scanner pair, the average wavelet score and Laplacian variance did not show statistically significant differences between the two groups and no significant association was observed between prediction concordance and the proportion of blurry regions (<em>p</em> > 0.05, except one pair). Model performance remained robust even at a high blur level (radius = 1), where the patch image had a Laplacian variance of 133.14 and a wavelet score of 1667.98, corresponding to the top 8.6% and 12.15% of blurriness, respectively, in our dataset. In addition, embedding analysis across focal planes using <em>Z</em>-stacks revealed that both patch- and slide-level representations were preserved up to ±3 μm. Slide-level embeddings consistently exhibited cosine similarity values above 0.99.</div></div><div><h3>Conclusions</h3><div>These findings empirically suggest that the typical levels of WSI blurriness encountered in clinical practice may not significantly compromise the robustness of slide-level AI classification.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"19 ","pages":"Article 100518"},"PeriodicalIF":0.0,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145320091","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Natural language processing for deep phenotyping of patients receiving genomic testing enables effective gene prioritization in a clinical diagnostics pipeline 对接受基因组检测的患者进行深度表型分析的自然语言处理能够在临床诊断管道中实现有效的基因优先排序

Q2 Medicine

Journal of Pathology Informatics

Pub Date : 2025-11-01 Epub Date: 2025-12-13 DOI: 10.1016/j.jpi.2025.100503

Andy Drackley , Anthony Wong , Patrick McMullen , Alexander Ing , Pamela Rathbun , Kai Lee Yap

引用次数: 0

Evaluating a patient matching algorithm in a Hematopathology Database (HPDB2): preliminary data analysis and comparison to an epic-based approach 评估血液病数据库（HPDB2）中的患者匹配算法：初步数据分析和与基于史诗的方法的比较

Q2 Medicine

Journal of Pathology Informatics

Pub Date : 2025-11-01 Epub Date: 2025-12-13 DOI: 10.1016/j.jpi.2025.100477

Matthew Xi Luo , Willow Solem , Niklas Krumm

引用次数: 0

Using a Git service provider and the web browser as an application server for clinical pathology job aids 采用Git服务提供程序和web浏览器作为应用服务器，实现临床病理作业辅助

Q2 Medicine

Journal of Pathology Informatics

Pub Date : 2025-11-01 Epub Date: 2025-12-13 DOI: 10.1016/j.jpi.2025.100481

Christopher Williams

引用次数: 0

From detecting variants to matching trials: assessing ChatGPT-4o's utility in molecular pathology 从检测变异到匹配试验：评估chatgpt - 40在分子病理学中的效用

Q2 Medicine

Journal of Pathology Informatics

Pub Date : 2025-11-01 Epub Date: 2025-12-13 DOI: 10.1016/j.jpi.2025.100475

Sidra Zaheer , Mine Yilmaz , Lahari Koganti , Mahesh M. Mansukhani , Susan J. Hsiao

引用次数: 0

Weakly supervised deep learning-based detection of serous tubal intraepithelial carcinoma in fallopian tubes 基于弱监督深度学习的输卵管浆液性上皮内癌检测

Q2 Medicine

Journal of Pathology Informatics

Pub Date : 2025-11-01 Epub Date: 2025-10-23 DOI: 10.1016/j.jpi.2025.100522

Andrew L. Valesano, Stephanie L. Skala , Mustafa Yousif

Serous tubal intraepithelial carcinoma (STIC) is an uncommon, non-invasive carcinoma that occurs more frequently in individuals with germline BRCA mutations and is an established precursor to high-grade serous ovarian carcinoma. STIC can be challenging to detect during pathologist evaluation, as it can manifest as a small focus of atypia in an otherwise benign salpingectomy specimen. There is a clinical need for scalable, weakly supervised computational approaches to aid in the detection of STIC. We developed a deep learning model to identify STIC and serous tubal intraepithelial lesions (STIL) in whole-slide images. We obtained fallopian tube specimens diagnosed as STIC (n = 49), STIL (n = 48), and benign fallopian tube (n = 83) at a single academic medical center. We trained a weakly supervised, attention-based multiple instance learning model and evaluated performance on independent datasets, including an additional unbalanced dataset (n = 40 benign, n = 2 STIL, n = 1 STIC) and cases diagnosed descriptively as benign reactive atypia (n = 53). The model achieved high sensitivity and specificity on the balanced validation cohort, with an area under the receiver operating characteristic curve (AUROC) of 0.96 (95% CI: 0.90–1.00), and demonstrated similarly strong performance on unbalanced validation cohorts (AUROC 0.98). Interpretability analyses indicated that model decisions were based on epithelial atypia. These results support the potential of integrating deep learning screening tools into clinical workflows to augment pathologist efficiency and diagnostic accuracy in fallopian tubes.

浆液性输卵管上皮内癌（STIC）是一种罕见的、非侵袭性的癌症，多发生于BRCA种系突变个体，是高级别浆液性卵巢癌的先兆。在病理评估中发现STIC是很有挑战性的，因为它可以在良性输卵管切除术标本中表现为一个小的异型灶。临床需要可扩展的、弱监督的计算方法来帮助检测STIC。我们开发了一个深度学习模型来识别全片图像中的STIC和浆液性输卵管上皮内病变（STIL）。我们在一个学术医疗中心获得诊断为STIC （n = 49）、STIL （n = 48）和良性输卵管（n = 83）的输卵管标本。我们训练了一个弱监督的、基于注意力的多实例学习模型，并在独立数据集上评估其性能，包括一个额外的不平衡数据集（n = 40个良性数据集，n = 2个STIL数据集，n = 1个STIC数据集）和被描述诊断为良性反应性非典型型的病例（n = 53）。该模型在平衡验证队列中具有很高的灵敏度和特异性，受试者工作特征曲线下面积（AUROC）为0.96 (95% CI: 0.90-1.00)，在不平衡验证队列中也表现出同样强的性能（AUROC为0.98）。可解释性分析表明，模型的决定是基于上皮异型性。这些结果支持将深度学习筛选工具整合到临床工作流程中，以提高输卵管病理学家的效率和诊断准确性。

{"title":"Weakly supervised deep learning-based detection of serous tubal intraepithelial carcinoma in fallopian tubes","authors":"Andrew L. Valesano, Stephanie L. Skala , Mustafa Yousif","doi":"10.1016/j.jpi.2025.100522","DOIUrl":"10.1016/j.jpi.2025.100522","url":null,"abstract":"<div><div>Serous tubal intraepithelial carcinoma (STIC) is an uncommon, non-invasive carcinoma that occurs more frequently in individuals with germline <em>BRCA</em> mutations and is an established precursor to high-grade serous ovarian carcinoma. STIC can be challenging to detect during pathologist evaluation, as it can manifest as a small focus of atypia in an otherwise benign salpingectomy specimen. There is a clinical need for scalable, weakly supervised computational approaches to aid in the detection of STIC. We developed a deep learning model to identify STIC and serous tubal intraepithelial lesions (STIL) in whole-slide images. We obtained fallopian tube specimens diagnosed as STIC (<em>n</em> = 49), STIL (<em>n</em> = 48), and benign fallopian tube (<em>n</em> = 83) at a single academic medical center. We trained a weakly supervised, attention-based multiple instance learning model and evaluated performance on independent datasets, including an additional unbalanced dataset (<em>n</em> = 40 benign, <em>n</em> = 2 STIL, <em>n</em> = 1 STIC) and cases diagnosed descriptively as benign reactive atypia (<em>n</em> = 53). The model achieved high sensitivity and specificity on the balanced validation cohort, with an area under the receiver operating characteristic curve (AUROC) of 0.96 (95% CI: 0.90–1.00), and demonstrated similarly strong performance on unbalanced validation cohorts (AUROC 0.98). Interpretability analyses indicated that model decisions were based on epithelial atypia. These results support the potential of integrating deep learning screening tools into clinical workflows to augment pathologist efficiency and diagnostic accuracy in fallopian tubes.</div></div>","PeriodicalId":37769,"journal":{"name":"Journal of Pathology Informatics","volume":"19 ","pages":"Article 100522"},"PeriodicalIF":0.0,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145525750","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

首页上一页

下一页尾页

类型

全部化学•材料生命科学医学物理工程技术环境•农林材料科学地球科学法学管理学化学环境科学与生态学计算机科学教育学经济学农林科学人文科学生物学数学物理与天体物理心理学综合性期刊其他工业工程理学历史学农学文学信息工程

数据库

全部 ACS Publications Elsevier ieeexplore Springer The Royal Society of Chemistry Wiley

期刊

Journal of Pathology Informatics

全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.

﹀