浏览全部资源
扫码关注微信
三明市中西医结合医院药学部,福建 三明 365001
Published:15 December 2023,
Received:23 May 2023,
Revised:15 September 2023,
扫 描 看 全 文
陈祺焘,倪璟雯,徐君等.生成式人工智能GPT-4驱动的中药处方生成研究 Δ[J].中国药房,2023,34(23):2825-2828.
CHEN Qitao,NI Jingwen,XU Jun,et al.Generation of traditional Chinese medicine prescription driven by generative artificial intelligence GPT-4[J].ZHONGGUO YAOFANG,2023,34(23):2825-2828.
陈祺焘,倪璟雯,徐君等.生成式人工智能GPT-4驱动的中药处方生成研究 Δ[J].中国药房,2023,34(23):2825-2828. DOI: 10.6039/j.issn.1001-0408.2023.23.02.
CHEN Qitao,NI Jingwen,XU Jun,et al.Generation of traditional Chinese medicine prescription driven by generative artificial intelligence GPT-4[J].ZHONGGUO YAOFANG,2023,34(23):2825-2828. DOI: 10.6039/j.issn.1001-0408.2023.23.02.
目的
2
评估生成式人工智能(AIGC)中的GPT-4模型生成中药处方的安全性、适宜性,为AIGC赋能中医药行业提供研究思路。
方法
2
将2020年版《中国药典》和第5版《中药学》作为语料,由GPT-4及基于GPT-4开发的实时联网模型(简称“联网模型”)对其进行深度学习。人工抽取近几年中医药类专家共识收录的临床案例,由GPT-4模型和联网模型根据诊断重新生成处方。由中医药学专家对GPT-4生成处方、联网模型生成处方以及专家共识处方进行盲评打分,同时通过图灵测试来评估GPT-4模型和联网模型是否具有与人类智能相当的能力。
结果
2
GPT-4模型生成的中药处方的平均分与人工处方比较,差异无统计学意义(
P
>0.05);联网模型生成处方的平均分与GPT-4模型生成的中药处方比较,差异无统计学意义(
P
>0.05)。模型生成处方在图灵测试中被误判为人工处方的占比达51.11%。
结论
2
GPT-4模型生成的中药处方在安全性、适宜性方面已经具备一定的水平,且GPT-4模型通过了所设置的图灵测试;在诊疗过程中引入AIGC可能为临床中药的合理使用提供技术支撑。
OBJECTIVE
2
To evaluate the safety and suitability of traditional Chinese medicine prescriptions generated by generative artificial intelligence (AIGC), and to provide research ideas for empowering the traditional Chinese medicine industry with AIGC.
METHODS
2
Using the 2020 edition of
Chinese Pharmacopoeia
and the 5th edition of
Traditional Chinese Medicine
as corpus, GPT-4 and the real-time networking model developed based on GPT-4 (referred to as the “networking model”) were used for deep learning. The clinical cases included in the consensus of traditional Chinese medicine experts in recent years were extracted manually to regenerate prescriptions based on diagnosis using the GPT-4 model and networking model; traditional Chinese medicine experts conducted blind evaluation and scoring of GPT-4 generated prescriptions, networking model generated prescriptions, and expert consensus prescriptions. At the same time, Turing testing was used to evaluate whether the GPT-4 model and networking model had the same ability as human intelligence.
RESULTS
2
The average score of traditional Chinese medicine prescriptions generated by the GPT-4 model showed no statistically significant difference compared to manual prescriptions (
P
>0.05), while the average score of prescriptions generated by the networking model showed no statistically significant difference compared to traditional Chinese medicine prescriptions generated by the GPT-4 model (
P
>0.05). The proportion of model-generated prescriptions mistakenly judged as manual prescriptions in the Turing test was 51.11%.
CONCLUSIONS
2
The traditional Chinese medicine prescriptions generated by the GPT-4 model have reached a certain level of safety and suitability, and the GPT-4 model has passed the Turing test. The introduction of AIGC in the diagnosis and treatment process may provide technical support for the rational use of clinical traditional Chinese medicine.
GPT-4中药处方生成式人工智能
traditional Chinese medicine prescriptiongenerative artificial intelligence
杨洪军,李耿. 推动中药产业迈向高质量发展[J]. 中国生物工程杂志,2022,42(5):16-17.
YANG H J,LI G. Promote the development of Chinese medicine industry towards high quality[J]. China Biotechnol,2022,42(5):16-17.
彭善祥. 中药处方点评在中药饮片合理使用中的干预价值分析[J]. 北方药学,2022,19(3):88-90.
PENG S X. Analysis of intervention value of traditional Chinese medicine prescription comment on rational use of traditional Chinese medicine decoction pieces[J]. J N Pharm,2022,19(3):88-90.
胡宗仁,邓奕辉,沈承玲,等. “西学中” 系统化培训的意义、特点及知识体系[J]. 中国中西医结合杂志,2022,42(12):1424-1427.
HU Z R,DENG Y H,SHEN C L,et al. Significance,cha- racteristics and knowledge system of systematic training of western medicine doctors learning from Chinese medicine[J]. Chin J Integr Tradit West Med,2022,42(12):1424-1427.
方邦江,张洪春,张忠德,等. 2023年春季成人流行性感冒中医药防治专家共识[J]. 陕西中医药大学学报,2023,46(4):1-5.
FANG B J,ZHANG H C,ZHANG Z D,et al. Expert consensus on traditional Chinese medicine prevention and treatment of adult influenza in spring 2023 [J]. J Shaanxi Univ Tradit Chin Med,2023,46(4):1-5.
孙增涛,师艺航,李小娟. 咳嗽中医诊疗专家共识意见:2021[J]. 中医杂志,2021,62(16):1465-1472.
SUN Z T,SHI Y H,LI X J.Experts consensus on the diagnosis and treatment of cough in traditional Chinese medicine:2021[J]. J Tradit Chin Med,2021,62(16):1465-1472.
王万春,严张仁. 毒蛇咬伤中医诊疗方案专家共识:2016版[J]. 中医杂志,2017,58(4):357-360.
WANG W C,YAN Z R. Expert consensus on TCM diagnosis and treatment scheme for snakebite:2016 edition[J]. J Tradit Chin Med,2017,58(4):357-360.
刘江峰,刘雏菲,齐月,等. AIGC助力数字人文研究的实践探索:SikuGPT驱动的古诗词生成研究[J]. 情报理论与实践,2023,46(5):23-31.
LIU J F,LIU C F,QI Y,et al. A practical exploration of AIGC-powered digital humanities research:a SikuGPT driven research of ancient poetry generation[J]. Inf Stud Theory Appl,2023,46(5):23-31.
刘堃. SPSS统计分析在医学科研中的应用[M]. 北京:人民卫生出版社,2012:1-244.
LIU K. Application of SPSS statistical analysis in medical scientific research[M]. Beijing:People’s Medical Publi- shing House,2012:1-244.
殷保才. 基于深度学习的医学影像辅助诊断技术研究[D]. 合肥:中国科学技术大学,2022.
YIN B C. Research on computer aided diagnosis of medical images based on deep learning[D]. Hefei:University of Science and Technology of China,2022.
陈吉栋,何丽. 生成式人工智能风险治理亟待“伦理-法律” 综合框架[J]. 张江科技评论,2023(2):8-10.
CHEN J D,HE L. Risk management of generative artificial intelligence urgently needs a comprehensive framework of “ethics-law”[J]. Zhangjiang Technol Rev,2023(2):8-10.
宋黎磊,戴淑婷. 科技安全化与泛安全化:欧盟人工智能战略研究[J]. 德国研究,2022,37(4):47-65,125.
SONG L L,DAI S T. Shuting technology security and pan security:research on EU artificial intelligence strategy[J] Ger Res,2022,37(4):47-65,125.
0
Views
7
下载量
0
CSCD
Publicity Resources
Related Articles
Related Author
Related Institution