摘要
深度学习是人工智能在机器学习领域的一大分支,其应用过程不需要做大量特定领域知识的特征提取便可以得到传统方式难以提取到的数据特征,完成复杂的数据分析,相对于传统机器学习更为高效准确。当前深度学习在中医学领域已有初步发展,在辅助中医诊断、构建中医诊断模型、医案数据处理等方面具备显著成果。本文基于深度学习在语音识别、数据图像处理及自然语言理解方面的优势,结合其在中医药研究中的应用现状,对深度学习的发展前景和当下存在问题进行思考讨论,认为在今后深度学习和中医药融合发展的过程中,应注意算法模型的性能水平及数据样本的数量和质量,为深度学习的发展提供规范的中医专业数据库是当前亟需解决的问题。
Deep learning is a branch of artificial intelligence in the field of machine learning,its application process does not need a lot of feature extraction of domain-specific knowledge to obtain data features that are difficult to extract in traditional ways.It can perform complex data analysis more efficiently and accurately than traditional machine learning.At present,deep learning has been preliminarily developed in the field of traditional Chinese medicine,and has achieved remarkable results in assisting traditional Chinese medicine diagnosis,constructing traditional Chinese medicine diagnosis model,and medical case data processing.Based on the advantages of deep learning in speech recognition,data image processing and natural language understanding,combined with its application status in traditional Chinese medicine research,this paper discusses the development prospects and current problems of deep learning.It is believed that in the future process of the integration of deep learning and traditional Chinese medicine,attention should be paid to the performance level of algorithm models and the quantity and quality of data samples.It is an urgent problem to provide a standardized traditional Chinese medicine professional database for the development of deep learning.
作者
王天琳
姚魁武
WANG Tianlin;YAO Kuiwu(Department of Cardiology,Guang’anmen Hospital,China Academy of Chinese Medical,Beijing 100053,China;Department of Internal Medicine,Ophthalmology Hospital,China Academy of Chinese Medicine Sciences,Beijing 100040,China)
出处
《中国医药导报》
CAS
2024年第6期193-196,共4页
China Medical Herald
基金
国家自然科学基金资助项目(81873173)
国家重点研发计划项目(2019YFC1708703)
中国中医科学院科技创新工程重大攻关项目(C12021A00906)
北京市自然科学基金面上项目(7232324)。
关键词
深度学习
中医学
机器学习
算法模型
大数据
Deep learning
Traditional Chinese Medicine
Machine learning
Algorithm model
Big data
