“数据-知识”双驱动模式的辨证决策方法探索——以胃癌前状态为例  

Exploration of Decision-Making Methods Based on Syndrome Differentiation by“Data-Knowledge”Dual-Driven Models:A Case Study of Gastric Precancerous State

作　　者：徐伟超 杜艳茹[2,5] 郎晓猛[2,5] 娄莹莹 贾文文[2,5] 康欣 郭硕[3] 张坤 苏春芝[2] 田军彪 魏晓娜[2] 杨倩 XU Weichao;DU Yanru;LANG Xiaomeng;LOU Yingying;JIA Wenwen;KANG Xin;GUO Shuo;ZHANG Kun;SU Chunzhi;TIAN Junbiao;WEI Xiaona;YANG Qian(Graduate School,Hebei University of Chinese Medicine,Shijiazhuang,050091;Hebei Provincial Hospital of Traditional Chinese Medicine;The Fourth Hospital of Hebei Medical University;School of Information Science and Engineering,Hebei University of Science and Technology;Key Laboratory of Turbidity Toxin Syndrome(Hebei))

机构地区：[1]河北中医药大学研究生学院,河北省石家庄市050091 [2]河北省中医院 [3]河北医科大学第四医院 [4]河北科技大学信息科学与工程学院 [5]河北省浊毒证重点实验室

出　　处：《中医杂志》2024年第2期154-158,共5页Journal of Traditional Chinese Medicine

基　　金：国家自然科学基金(82205314);国家中医药管理局全国老中医药专家学术经验继承工作项目;河北省重点研发计划(21377724D,23377701D);河北省“三三三人才工程”资助项目(C20231017)。

摘　　要：数据分析模型可辅助中医药经验的传承和临床诊疗,以胃癌前状态为例探索“数据-知识”双驱动模式构建的可能性。数据驱动是围绕数据分析开展临床决策,其辨证决策研究依托隐结构模型、部分可观察马尔可夫决策过程以确认疾病病因、证候要素、病机演变和辨证方案;知识驱动是利用数据和信息推动决策和行动的过程,其辨证决策研究依托卷积神经网络以提高局部辨病辨证准确性。“数据-知识”双驱动模式可弥补单一驱动数字模拟准确性的短板,达到局部辨病与宏观辨证的平衡。在既往研究基础上探索胃癌前状态病证诊断辅助决策平台的构建方法,认为可通过机器与算法辅助来延伸医生诊断和决策能力,整合相关研究方法并获取胃癌前状态中医辨证和内镜病理诊断预测核心特征,探索疾病中医辨证内镜病理识别要素,为数据分析前沿技术在中医辨证智能化领域的深入研究和创新应用提供思路。agnosis and treatment,and the possibility of constructing a“data-knowledge”dual-drive model was explored by taking gastric precancerous state as an example.Data-driven is to make clinical decisions around data analysis,and its syndrome-differentiation decision-making research relies on hidden structural models and partially observable Markov decision-making processes to identify the etiology of diseases,syndrome elements,evolution of pathogenesis,and syndrome differentiation protocols;knowledge-driven is to make use of data and information to promote decision-making and action processes,and its syndrome-differentiation decision-making research relies on convolutional neural networks to improve the accuracy of local disease identification and syndrome differentiation.The“data-knowledge”dual-driven model can make up for the shortcomings of single-drive numerical simulation accuracy,and achieve a balance between local disease identification and macroscopic syndrome differentiation.On the basis of previous research,we explored the construction method of diagnostic assisted decision-making platform for gastric precancerous state,and believed that the diagnostic and decision-making ability of doctors can be extended through the assistance of machines and algorithms.Meanwhile,the related research methods were integrated and the core features of gastric precancerous state based on TCM syndrome differentiation and endoscopic pathology diagnosis and prediction were obtained,and the elements of endoscopic pathology recognition based on TCM syndrome differentiation were explored,so as to provide ideas for the in-depth research and innovative application of cutting-edge data analysis technology in the field of intelligent TCM syndrome differentiation.

关 键 词：胃癌前状态 数据分析模型 数据驱动 知识驱动 辨证决策 

分 类 号：R273[医药卫生—中西医结合]