摘要
基于近几年机器视觉的发展,深度学习的人工智能方法应用于组织病理极大程度上促进了病理学家解决临床上的诊断问题,用该种方法解决病理学问题可被称为计算机病理学。人工智能可以做到帮助病理学家初筛大部分良性数据、辅助诊断、疗效预测、识别生物标志物等,甚至可以做到对药效治疗监测以及识别药物发现未知的信号。基于深度学习在病理领域的深入研究,让计算机自动处理病理数据成为可能。人工智能诊断决策建立在大数据之上,很多有可能做到对每个病人的个性化管理,对于大多普遍性的疾病诊断有着更加快速准确的优势。但数字病理学的发展仍受到一些问题的限制,以至于现阶段没有广泛应用于数字病理诊断平台。本文总结了近几年人工智能在病理诊断领域的最新进展,并讨论这种技术的可行性,补充说明在数字病理学中遇到的困难和挑战,并提出在该领域实用性上的展望。
Based on the development of machine vision in recent years,the application of deep learning artificial intelligence methods in histopathology has greatly promoted pathologists to solve clinical diagnostic problems.This method can be called computer pathology.Artificial intelligence can help pathologists sift through most benign data,aid in diagnosis,predict efficacy,identify biomarkers,and even monitor therapeutic efficacy and identify unknown signals for drugs.Based on the indepth study of deep learning in the field of pathology,it is possible for the computer to process pathological data automatically.Artificial intelligence diagnostic decisions are based on big data,and it is possible to personalize the management of each patient.It has the advantage of more rapid and accurate diagnosis for most common diseases.However,it is worth considering that the development of digital pathology is still limited by some problems,so that it is not widely used in digital pathology diagnostic platform at the present stage.We summarized the recent progress of artificial intelligence in the field of pathological diagnosis in this paper.We discussed the feasibility of this technology,added the difficulties and challenges encountered in digital pathology,and put forward the prospect of its practicality in this field.
作者
余净纯
郭明星
韩靖
张小鹰
陈汉威
王浩
YU Jingchun;GUO Mingxing;HAN Jing;ZHANG Xiaoying;CHEN Hanwei;WANG Hao(South China Normal University-Panyu Central Hospital Joint Laboratory of Basis and Translational Medical Research,Guangzhou 511400,China;College of Life Sciences,South China Normal University,Guangzhou 510630,China;Department of Pathology,Guangzhou Panyu District Central Hospital,Guangzhou 511400,China)
出处
《分子影像学杂志》
2022年第5期779-789,共11页
Journal of Molecular Imaging
基金
国家自然科学基金海外及港澳学者延续重点资助项目(81729003)。
关键词
数字病理学
计算机辅助诊断
深度学习
全切片图像
digital pathological
computer-aided diagnosis
deep learning
whole slide image
