摘要
毒性病理学是促进动物和人类健康发展最有价值的学科之一,药物非临床安全性评价毒性研究中对石蜡包埋、苏木精和伊红染色切片的组织病理学检查是毒性病理学评价的金标准。数字毒性病理学、人工智能(artificial intelligence,AI)尤其是机器学习(machine learning,ML)是全球颠覆性、快速发展的技术领域,其对组织病理学领域的影响正在迅速显现。组织病理学检查种类繁多算法的发展和应用,表明人工智能病理学平台可深度影响将来数字毒性病理学、精准医疗和个性化医疗。然而,与所有其他革命性的技术相同,人工智能病理学平台在实施和应用过程中存在诸多挑战。本文综述了人工智能和机器学习的发展、人工智能在毒性病理学中的应用、机器学习在数字毒性病理学中的应用以及人工智能对数字毒性病理学的影响,以期为我国毒性病理学中人工智能和机器学习的应用提供一定参考。
Toxicologic pathology is one of the most valuable disciplines contributing to the advancement of animal and human health.The gold standard of the toxicologic pathology evaluation in toxicity studies during nonclinical safety evaluation of drugs is considered to be the histopathological examination of paraffin-embedded,hematoxylin and eosin-stained tissue sections.Digital toxicologic pathology,artificial intelligence(Al),and in particular machine learning(ML)are globally disruptive,rapidly growing sectors of technology whose impact on the field of histopathology is quickly being realized.The development and application of increasing numbers of algorithms in the histopathological field have demonstrated that Al pathology platforms are now poised to truly impact the future of digital toxicologic pathology,precision medicine,and personalized medicine.However,as with all great technological advances,there are implementation and adoption challenges.The development of AI and ML,application of AI in toxicologic pathology,application of ML in digital toxicologic pathology,and impact of AI on digital toxicologic pathology were reviewed in the paper,in order to provide some references for applying Al and ML in toxicologic pathology in China.
作者
李一昊
滕伊洋
张亚群
钱庄
胡文元
钟小群
胡静
陈晓俊
闫振龙
彭瑞楠
王娅
李慧
葛建雅
缪成贤
邵薇
吕建军
大平东子
LI Yi-hao;TENG Yi-yang;ZHANG Ya-qun;QIAN Zhuang;HU Wen-yuan;ZHONG Xiao-qun;HU Jing;CHEN Xiao-jun;YAN Zhen-long;PENG Rui-nan;WANG Ya;LI Hui;CE Jian-ya;MIAO Cheng-xian;SHAO Wei;LV Jian-jun;OHIRA Toko(InnoStar Bio-Tech Nantong Co.,Ltd.,Nantong 226133,China;Shanghai InnoStar Bio-Tech Co.,Ltd.,Shanghai 200043,China;China State Institute of Pharmaceutical Industry,Shanghai 201203,China)
出处
《中国新药杂志》
CAS
CSCD
北大核心
2023年第6期598-604,共7页
Chinese Journal of New Drugs
基金
干细胞治疗产品的规范化与规模化生产及质量评价研究(G2021086002L)。
关键词
毒性病理学
人工智能
机器学习
深度学习
人工神经网络
toxicologic pathology
artificial intelligence
machine learning
deep learning
artificial neural network