智能化挖掘中医临床诊疗数据面临的问题和挑战

Problems and challenges in intelligently mining TCM clinical diagnosis and treatment data

中医学以其独特的诊疗技术和明确的疗效,在服务全民健康中做出了重要的贡献,但面对巨大的全民健康需求,现有中医诊疗模式的服务能力和质量存在的问题愈发明显。互联网与人工智能技术的迅速发展,为中医诊疗模式的升级优化提供了新的机遇,在中医诊疗模式智能化进程中,对中医临床诊疗数据的挖掘处理技术是中医智能化的关键环节,也愈发成为限制中医智能化的瓶颈问题。究其原因,中医临床数据载体虽然丰富,但难以实现汇交融合;定性描述为主,难以客观量化处理;显隐知识复合,难以展示思维过程;诊疗体系自洽,难以构建通用模型。提高数据汇交融合和客观量化程度,探索隐性知识显化和多模态模拟技术,是提高中医诊疗数据智能分析和应用水平的重要着力点。

With its unique diagnosis and treatment technology and clear curative effect,Chinese medicine has made an important contribution to serving the health of the whole people.However,in the face of huge national health needs,the existing problems in the service capacity and quality of the existing Chinese medicine diagnosis and treatment model have become more and more obvious.The rapid development of the Internet and artificial intelligence technology has provided new opportunities for the upgrading and optimization of the Chinese medicine diagnosis and treatment model.In the process of the intelligentization of the Chinese medicine diagnosis and treatment mode,the mining and processing technology of the Chinese medicine clinical diagnosis and treatment data is the key link in the intelligentization of the Chinese medicine.Development has become a bottleneck that limits the intellectualization of Chinese medicine.The reason is that although TCM clinical data carriers are abundant,it is difficult to achieve convergence and integration;Qualitative description is mainly used,and it is difficult to objectively and quantify processing;Explicit tacit knowledge is complex,which makes it difficult to show the thinking process;The diagnosis and treatment system is self-consistent,and it is difficult to construct a universal model.Improving the degree of data collection,integration and objective quantification,and exploring tacit knowledge manifestation and multi-modal simulation technologies are important points for improving the level of intelligent analysis and application of Chinese medicine diagnosis and treatment data.