摘要
人工智能(artificial intelligence,AI)已用于肾脏病学的众多领域,将帮助医师优化临床诊疗、血液透析处方、肾脏组织病理学和移植患者管理。在肾脏病理方面,AI通过深度学习模型实现肾小球与肾小管间质病变的精准识别与量化,并基于此预测疾病进展;在临床诊疗中,AI模型在急性肾损伤早期预警、IgA肾病与慢性肾脏病预后评估,以及血液透析的过程管理和合并症治疗中展现出重要价值。此外,AI还应用于预测移植肾存活、辅助罕见病筛查及加速药物研发等场景。然而,许多肾脏科医师仍然不熟悉医疗AI的基本原理,且其在数据隐私、算法透明度及伦理监管等方面仍面临挑战。本综述旨在概述AI在肾脏病学各个领域中的应用进展,并对其发展前景与待解决问题进行展望。
Artificial intelligence(AI)has been applied across numerous fields of nephrology,assisting physicians in optimizing clinical diagnosis and treatment,hemodialysis prescriptions,renal histopathology,and transplant patient management.In renal pathology,AI utilizes deep learning models to achieve precise identification and quantification of glomerular and tubulointerstitial lesions,thereby predicting disease progression.In clinical practice,AI models demonstrate significant value in the early warning of acute kidney injury,prognosis assessment of IgA nephropathy and chronic kidney disease,as well as in the management of hemodialysis processes and comorbidity treatment.Furthermore,AI is also applied in predicting transplant kidney survival,assisting in the screening of rare diseases,and accelerating drug development.However,many nephrologists remain unfamiliar with the fundamental principles of medical AI,and challenges persist in areas such as data privacy,algorithm transparency,and ethical regulation.This review aims to outline the advances in the application of AI across various domains of nephrology and to discuss its future prospects and unresolved issues.
作者
肖雨
周阳
Xiao Yu;Zhou Yang(Center for Kidney Disease,the Second Affiliated Hospital of Nanjing Medical University,Nanjing 210003,China)
出处
《中华肾脏病杂志》
北大核心
2025年第11期880-888,共9页
Chinese Journal of Nephrology
基金
江苏省自然科学基金(BK20201497)。
关键词
肾脏病学
人工智能
深度学习
肾脏终点
肾脏病理
电子健康记录
Nephrology
Artificial intelligence
Deep learning
Kidney outcome
Renal pathology
Electronic health records
