[1]肖建力,许东舟,王浩,等.医疗领域的大型语言模型综述[J].智能系统学报,2025,20(3):530-547.[doi:10.11992/tis.202405003]
XIAO Jianli,XU Dongzhou,WANG Hao,et al.Survey of large language models in healthcare[J].CAAI Transactions on Intelligent Systems,2025,20(3):530-547.[doi:10.11992/tis.202405003]
点击复制
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
20
期数:
2025年第3期
页码:
530-547
栏目:
综述
出版日期:
2025-05-05
- Title:
-
Survey of large language models in healthcare
- 作者:
-
肖建力1, 许东舟1, 王浩2, 刘敏3, 周雷4, 朱林4, 顾松5
-
1. 上海理工大学 光电信息与计算机工程学院, 上海 200093;
2. 上海交通大学医学院附属上海儿童医学中心 心胸外科, 上海 200127;
3. 复旦大学附属妇产科医院 中西医结合妇科, 上海 200011;
4. 上海理工大学 健康科学与工程学院, 上海 200093;
5. 上海市第一人民医院 创伤临床医学中心, 上海 201620
- Author(s):
-
XIAO Jianli1, XU Dongzhou1, WANG Hao2, LIU Min3, ZHOU Lei4, ZHU Lin4, GU Song5
-
1. School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China;
2. Department of Cardiothoracic Surgery, Shanghai Children’s Medical Center, School of Medicine, Shanghai Jiao Tong University, Shanghai 200127, China;
3. Department of Gynecology of Integrated Traditional Chinese and Western Medicine, Obstetrics and Gynecology Hospital of Fudan University, Shanghai 200011, China;
4. School of Health Science and Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China;
5. Trauma Center, Shanghai General Hospital, Shanghai 201620, China
-
- 关键词:
-
人工智能; 深度学习; Transformer; 大型语言模型; 智慧医疗; 数据分析; 图像处理; 计算机视觉
- Keywords:
-
artificial intelligence; deep learning; Transformer; large language model; intelligent healthcare; data analysis; image processing; computer vision
- 分类号:
-
TP18
- DOI:
-
10.11992/tis.202405003
- 摘要:
-
深度学习是人工智能领域的热门研究方向之一,它通过构建多层人工神经网络模仿人脑对数据的处理机制。大型语言模型(large language model,LLM)基于深度学习的架构,在无需编程指令的情况下,能通过分析大量数据以获得理解和生成人类语言的能力,被广泛应用于自然语言处理、计算机视觉、智慧医疗、智慧交通等诸多领域。文章总结了LLM在医疗领域的应用,涵盖了LLM针对医疗任务的基本训练流程、特殊策略以及在具体医疗场景中的应用。同时,进一步讨论了LLM在应用中面临的挑战,包括决策过程缺乏透明度、输出准确性以及隐私、伦理问题等,随后列举了相应的改进策略。最后,文章展望了LLM在医疗领域的未来发展趋势,及其对人类健康事业发展的潜在影响。
- Abstract:
-
Deep learning (DL) is a popular research area in artificial intelligence. It simulates the data processing mechanism of the human brain by constructing multilayer artificial neural networks. Large language models (LLMs) based on the DL architecture can understand and generate human language by analyzing enormous data without programming instructions. Thus, LLMs are widely employed in various domains, such as natural language processing, computer vision, intelligent healthcare, and intelligent transportation. This article summarizes the application of LLMs in the healthcare sector, exploring their basic training processes, specific strategies for executing healthcare tasks, and their applications in specific healthcare scenarios. It also discusses the challenges of applying LLMs to the healthcare field, including the lack of transparency in decision-making processes, the accuracy of the output contents, and issues related to privacy and ethics. Thereafter, several strategies for addressing these issues are discussed. Finally, the future development trends of LLM in healthcare, as well as its criticality in promoting human health, are discussed.
备注/Memo
收稿日期:2024-5-5。
基金项目:国家自然科学基金项目(61603257).
作者简介:肖建力,副教授,主要研究方向为人工智能与大数据。2023年吴文俊人工智能科学技术奖科技进步奖(科普项目)获得者,中国计算机学会杰出会员。发表学术论文10篇,著有图书《人工智能怎么学》。E-mail:audyxiao@sjtu.edu.cn。;许东舟,硕士研究生,主要研究方向为智慧医疗。E-mail:233370870@st.usst.edu.cn。;王浩,副主任医师,主要研究方向为先天性心脏病和先天性气管狭窄的外科治疗。E-mail: haowang_nt@163.com。
通讯作者:肖建力. E-mail:audyxiao@sjtu.edu.cn
更新日期/Last Update:
1900-01-01