Large Language Models in Psychiatry:Current Applications,Limitations,and Future Scope  

作　　者：Zhe Liu Yihang Bao Shuai Zeng Ruiyi Qian Miaohan Deng An Gu Jianye Li Weidi Wang Wenxiang Cai Wenhao Li Han Wang Dong Xu Guan Ning Lin 

机构地区：[1]Shanghai Mental Health Center,Shanghai Jiao Tong University School of Medicine,and School of Biomedical Engineering,Shanghai Jiao Tong University,Shanghai 200030,China [2]Department of Electrical Engineering and Computer Science,University of Missouri,Columbia,MO 65211,USA [3]School of Information Science and Technology,Institute of Computational Biology,Northeast Normal University,Changchun 130024,China

出　　处：《Big Data Mining and Analytics》2024年第4期1148-1168,共21页大数据挖掘与分析(英文)

基　　金：supported by STI 2030—Major Projects(No.2022ZD0209100);the National Natural Science Foundation of China(Nos.81971292,82150610506,and 62372099);the Natural Science Foundation of Shanghai(No.21ZR1428600);The Medical-Engineering Cross Foundation of Shanghai Jiao Tong University(No.YG2022ZD026);the Jilin Scientific and Technological Development Program(Nos.20230401092YY and 20230201090GX);the Paul K.and Diane Shumaker Endowment Fund at University of Missouri,USA.

摘　　要：With the advancements in Artificial Intelligence(AI)technology,Large Language Models(LLMs)provide outstanding capabilities for natural language understanding and generation,enhancing various domains.In psychiatry,LLMs can empower healthcare by analyzing vast amounts of medical data to improve diagnostic accuracy,enhance therapeutic communication,and personalize patient care with their strength in understanding and generating human-like text.In clinical AI,developing and utilizing robust and interpretable models has been a longstanding challenge.This survey investigates the current psychiatric practice of LLMs,along with a series of corpus resources that could be used for training psychiatric LLMs.We discuss the limitations concerning LLM reproducibility,capabilities,usability,interpretability in clinical settings,and ethical considerations.Additionally,we propose potential future directions for research,clinical application,and education in psychiatric LLMs.Finally,we discuss the challenge of integrating LLMs into the evolving landscape of healthcare in real-world scenarios.

关 键 词：Artificial Intelligence(AI) Large Language Model(LLM) PSYCHIATRY medical application 

分 类 号：TP368.33[自动化与计算机技术—计算机系统结构] TN929.53[自动化与计算机技术—计算机科学与技术]