一种基于计算机视觉的针刺手法分类系统开发与应用  

Development and application of computer vision-based acupuncture manipulation classification system

作　　者：涂涛 苏业豪 宿翀[1] 王磊 赵亚楠 陈捷 TU Tao;SU Ye-hao;SU Chong;WANG lei;ZHAO Ya-nan;CHEN Jie(College of Information Science and Technology,Beijing University of Chemical Technology,Beijing 100029,China;Institute of Acupuncture and Moxibustion,China Academy of Chinese Medical Sciences,Beijing 100700;Department of Traditional Chinese Medicine,Beijing Zhongguancun Hospital,Beijing 100190)

机构地区：[1]北京化工大学信息科学与技术学院,北京100029 [2]中国中医科学院针灸研究所,北京100700 [3]北京市中关村医院中医综合科,北京100190

出　　处：《针刺研究》2021年第6期469-473,共5页Acupuncture Research

基　　金：国家自然科学基金项目(No.82074285);北京市中医管理局科技发展资金项目(No.JJ-2020-07)。

摘　　要：目的:为了提升针刺手法建模与传承的准确性,面向针刺实践中的手法视频,本文探讨利用计算机视觉技术对中医针灸学中“捻转”和“提插”这两类基本针刺手法进行分类的可行性。方法:构建一种计算机视觉下的基于三维卷积神经网络和长短时记忆网络的混合深度学习网络模型,提取针刺手法视频帧序列的时空特征,将其输入分类器中实现分类。结果:针对200组录制的医师针刺手法视频,应用所提混合网络模型对“捻转”和“提插”两类手法进行分类,训练准确率达到95.4%,验证准确率达到95.3%。结论:本系统可为针刺手法的数据提取与传承提供一条有效途径。Objective To improve the accuracy of acupuncture manipulation modeling and inheritance,this article explores the feasibility of automatically classifying“twirling”and“lifting and thrusting”,two basic acupuncture manipulations in science of acupuncture and moxibustion,with the computer vision technology.Methods A hybrid deep learning network model was designed based on 3Dconvolutional neural network and long-short term memory neural network to extract the spatial-temporal features of video frame sequences,which were then input into the classifier for classification.Results The model discriminated between“twirling”and“lifting and thrusting”manipulations in 200videos,with the training and verification accuracy reaching up to 95.4%and 95.3%,respectively.Conclusion This computer vision-based acupuncture manipulation classification system provides an effective way for the data extraction and inheritance of acupuncture manipulations.

关 键 词：计算机视觉 针刺手法 三维卷积神经网络 长短时记忆网络 深度学习 

分 类 号：R245.31[医药卫生—针灸推拿学]