Assessing interstitial fibrosis and tubular atrophy in kidney biopsies - artificial intelligence versus humans.

IF 3.1 3区医学 Q3 PERIPHERAL VASCULAR DISEASE

Current Opinion in Nephrology and Hypertension Pub Date : 2026-05-01 Epub Date: 2026-03-13 DOI:10.1097/MNH.0000000000001174

Alton B Farris, Dženan Zukić, Kim Solez

{"title":"Assessing interstitial fibrosis and tubular atrophy in kidney biopsies - artificial intelligence versus humans.","authors":"Alton B Farris, Dženan Zukić, Kim Solez","doi":"10.1097/MNH.0000000000001174","DOIUrl":null,"url":null,"abstract":"Purpose of review: The degree to which computerized methods, such as artificial intelligence (AI), will aid in the assessment of kidney histopathology is undergoing intense study and application; and this is particularly true for interstitial fibrosis, which is often used as a surrogate measure of chronic kidney disease progression, since interobserver variability among human pathologists has been demonstrated in the assessment of interstitial fibrosis and other features.Recent findings: Computerized assessment of interstitial fibrosis, including with AI, has been assessed alongside pathologists. Computerized methods such as AI have shown direct interstitial fibrosis measurement and indirect assessment through kidney compartment segmentation; however, some studies have shown lack of complete concordance among computerized methods and humans; and studies have still shown the persistent value of human assessment in many circumstances.Summary: Computerized methods, including AI, are showing increased application in kidney pathology for a wide variety of clinical and histopathologic parameter assessment, including interstitial fibrosis; however, further studies are needed to characterize the performance of AI and handcrafted methods; and additional work is needed to fully integrate computerized methods into routine pathology practice. Ultimately, humans working with AI (\"humans + AI\") may provide enhanced analysis for more effective patient care.","PeriodicalId":10960,"journal":{"name":"Current Opinion in Nephrology and Hypertension","volume":" ","pages":"300-307"},"PeriodicalIF":3.1000,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Current Opinion in Nephrology and Hypertension","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1097/MNH.0000000000001174","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2026/3/13 0:00:00","PubModel":"Epub","JCR":"Q3","JCRName":"PERIPHERAL VASCULAR DISEASE","Score":null,"Total":0}

引用次数: 0

Abstract

Purpose of review: The degree to which computerized methods, such as artificial intelligence (AI), will aid in the assessment of kidney histopathology is undergoing intense study and application; and this is particularly true for interstitial fibrosis, which is often used as a surrogate measure of chronic kidney disease progression, since interobserver variability among human pathologists has been demonstrated in the assessment of interstitial fibrosis and other features.

Recent findings: Computerized assessment of interstitial fibrosis, including with AI, has been assessed alongside pathologists. Computerized methods such as AI have shown direct interstitial fibrosis measurement and indirect assessment through kidney compartment segmentation; however, some studies have shown lack of complete concordance among computerized methods and humans; and studies have still shown the persistent value of human assessment in many circumstances.

Summary: Computerized methods, including AI, are showing increased application in kidney pathology for a wide variety of clinical and histopathologic parameter assessment, including interstitial fibrosis; however, further studies are needed to characterize the performance of AI and handcrafted methods; and additional work is needed to fully integrate computerized methods into routine pathology practice. Ultimately, humans working with AI ("humans + AI") may provide enhanced analysis for more effective patient care.

查看原文本刊更多论文

评估肾活检间质纤维化和肾小管萎缩人工智能与人类。

综述目的：计算机化方法，如人工智能（AI），将在多大程度上帮助评估肾脏组织病理学正在进行激烈的研究和应用；对于间质性纤维化尤其如此，间质性纤维化常被用作慢性肾脏疾病进展的替代指标，因为人类病理学家在评估间质性纤维化和其他特征时已证实了观察者之间的差异。最近发现：与病理学家一起评估了间质纤维化的计算机评估，包括AI。计算机化方法如AI显示直接间质纤维化测量和通过肾隔室分割间接评估；然而，一些研究表明，计算机方法和人类之间缺乏完全的一致性；研究仍然表明，在许多情况下，人类评估的持续价值。摘要：计算机化方法，包括人工智能，在肾脏病理中越来越多地应用于各种临床和组织病理学参数评估，包括间质纤维化；然而，需要进一步的研究来表征人工智能和手工制作方法的性能；将计算机化方法完全整合到常规病理实践中还需要做更多的工作。最终，人类与人工智能合作（“人类+人工智能”）可能会为更有效的患者护理提供增强的分析。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Current Opinion in Nephrology and Hypertension 医学-泌尿学与肾脏学

CiteScore

5.70

自引率

6.20%

发文量

132

审稿时长

6-12 weeks

期刊介绍： A reader-friendly resource, Current Opinion in Nephrology and Hypertension provides an up-to-date account of the most important advances in the field of nephrology and hypertension. Each issue contains either two or three sections delivering a diverse and comprehensive coverage of all the key issues, including pathophysiology of hypertension, circulation and hemodynamics, and clinical nephrology. Current Opinion in Nephrology and Hypertension is an indispensable journal for the busy clinician, researcher or student.