首页|中医智能辨证多决策模型构建思路与方法

中医智能辨证多决策模型构建思路与方法

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
辨证受医师诊疗经验影响存在复杂性、模糊性、不确定性等弊端，融入人工智能技术进行智能化辨证是解决这些弊端的重要方法。但当前智能辨证面临使用模型单一、辨证模式难以适用于多病种等问题，使得辨证的准确度与适用度均有待提高。为解决智能化辨证当前存在的复杂问题，将提取文本信息、优化医案数据结构及模型选择与联合构建等作为多决策模型思路构建的重要组成部分，并根据数据和模型特点，采用自然语言处理技术进行医案症状及证型内容信息提取，基于粗糙集的属性约简算法进行数据降维处理，最后采取加权投票融合支持向量机、多标记K-近邻、反向传播(Back Propagation，BP)神经网络算法构建多决策辨证模型，旨在为提升人工智能辨证模型的准确率提供参考，更好地指导临床辨证。
Thinking and Method of Constructing Multi-Decision Model of Intelligent Syndrome Differentiation of Traditional Chinese Medicine
Syndrome differentiation is affected by doctors'experience in diagnosis and treatment,which has some drawbacks such as complexity,fuzziness and uncertainty.Integrating artificial intelligence technology into intelligent syndrome differentiation is an important method to solve these drawbacks.However,the current intelligent syndrome differentiation is faced with the prob-lems of single use model and syndrome differentiation model is difficult to apply to multiple diseases,which makes the accuracy and applicability of syndrome differentiation need to be improved.In order to solve the current complex problems in intelligent syndrome differentiation,this paper takes extracting text information,optimizing medical case data structure,model selection and joint construction as an important part of the multi decision model idea construction.According to the characteristics of data and models,natural language processing technology is used to extract medical case symptoms and syndrome type content information,and attribute reduction algorithm based on rough set is used to reduce data dimensions,finally,the weighted vote fusion support vector machine,multi-label K-nearest neighbor and back propagation(BP)neural network algorithm are adopted to build a multi-decision dialectical model,aiming to provide reference for improving the accuracy of artificial intelligence syndrome differ-entiation model and better guide clinical syndrome differentiation.

artificial intelligencemachine learningmodelsyndrome differentiation and treatment

展志宏、戴国华、张丛惠、任丽丽、管慧、高武霖

展开 >

山东中医药大学第一临床医学院,山东济南 250014

山东中医药大学附属医院,山东济南 250014

人工智能 机器学习 模型 辨证论治

国家重点研发计划中医药现代化研究重点专项国家自然科学基金面上项目国家自然科学基金青年科学基金项目中国博士后科学基金面上项目山东省自然科学基金青年项目

2019YFC17104018217141415822049422022M721998ZR2022QH123

2024

10.13193/j.issn.1673-7717.2024.02.004

中华中医药学刊
中华中医药学会 ,辽宁中医药大学

中华中医药学刊

CSTPCD北大核心
影响因子：1.007
ISSN：1673-7717
年,卷(期)：2024.42(2)
  • 37