北京生物医学工程

基于痧象的脏腑功能失调预警方法

The early warning method of visceral dysfunction based on Sha Xiang

作者：韩立博胡广芹张新峰张艳阳

单位：北京工业大学信息学部(北京100124) 北京工业大学生命科学与生物工程学院(北京100124)

关键词：中医诊断；痧象分析；图像处理；模式识别；脏腑功能

分类号：R318.04

出版年·卷·期（页码）：2021·40·3（233-238）

摘要：

目的在痧象临床研究中，医生可以通过观察背部脏腑穴位区域的痧象进行脏腑功能失调预警。传统的脏腑功能失调预警主要基于医生对背部脏腑穴位区域模糊定位来实现，面对较多的区域，无论是定位还是识别都需要花费很多时间，影响工作效率。利用相机采集的图像有一些杂乱的背景，影响痧象图像在电子病历中的规范化存储。针对此问题，本文提出一种基于图像处理与模式识别技术的痧象分析方法，进行脏腑功能失调预警，辅助医生进行诊断以及规范化存储痧象图像。方法采用基于两点的grabcut算法进行背部图像的校正、分割、脏腑穴位区域定位，分别利用支持向量机（support vector machine, SVM）、Alexnet、Alexnet算法基础上修改的卷积神经网络（convolutional neural networks, CNN）对脏腑穴位区域进行痧象识别。结果基于两点的grabcut算法可以实现背部图像的校正、分割以及脏腑穴位区域定位。相比于SVM、Alexnet，在Alexnet算法基础上修改后的CNN对痧象识别率最高，达到了99.83%，可进行痧象识别，从而快速进行脏腑功能失调预警。结论该方法对进行脏腑功能失调预警具有一定的实际应用价值。

Objective In the clinical study of Sha Xiang，doctor can observe the signs of Sha Xiang in the acupoint area of the back viscera for early warning of visceral dysfunction.The traditional early warning of visceral dysfunction requires doctor to make fuzzy location of the back viscera acupoint areas.In the face of more areas, both location and recognition need to spend a lot of time, which affects the work efficiency.The image collected by camera has some disordered background, which affects the standardized storage of Sha Xiang image in electronic medical record.In order to solve this problem, this paper proposes a method of image analysis based on image processing and pattern recognition technology, which is used for early warning of visceral dysfunction.It can assist doctor in diagnosis and standardized storage image of Sha Xiang.Methods Grabcut algorithm based on two points is used for back image correction, segmentation and locating the viscera acupoint area.Support vector machine (SVM), Alexnet and modified convolutional neural networks(CNN) based on Alexnet are used to identify the Sha Xiang in the acupoint area of the viscera.Results Grabcut algorithm based on two points can realize the back image correction , segmentation and locating the viscera acupoint area very well.Compared with SVM and Alexnet,modified convolutional neural networks based on Alexnet has the highest recognition rate of Sha Xiang.The recognition accuracy is 99.83%.It is very good for the recognition of Sha Xiang，so as to quickly carry out the early warning of visceral dysfunction.Conclusions This method has a certain practical value for the early warning of visceral dysfunction.

参考文献：

[1] 胡广芹,邓益群.痧疗帮您摆脱亚健康[J].中医健康养生,2018,4(12):30-31

Hu GQ,?Deng YQ.?Sha therapy helps you get rid?of sub-health[J].TCM Healthy Life-Nurturing,

2018,4(12):30-31

[2] 国家中医药管理局职业技能鉴定指导中心，国家中医药行业特有工种职业技能鉴定工作中医刮痧师职业专家委员会.中医刮痧师 [M].2版.北京：中国中医药出版社，2011.

[3]武文强,张新峰,孙艳玲,等.基于颜色与纹理的痤疮痧象证型自动分类初探[J].北京生物医学工程2014,33(6):565-569.

Wu WQ, Zhang XF , Sun YL , et al. Preliminary study of automatic classification for Sha image based on color and texture[J]. Beijing Biomedical Engineering, 2014,33(6):565-569.

[4]徐衍鲁.基于颜色特征的图像检索技术综述[J].电脑知识与技术,2017,13(13):156-158，163.

XU YL.?Image Retrieval Based on Color Feature[J].?Computer Knowledge and Technology,2017,

13(13):156-158，163.

[5]Rother C, Kolmogorov V ,Blake A. GrabCut: interactive foreground extraction using iterated graph cuts[J]. ACM Transactions on Graphics.2004, 23（3）:309-314.

[6]孙艳玲.中医痧象自动分析及其应用初步研究[D].北京：北京工业大学,2013.

Sun YL.Preliminary research on the automatic analysis and?application of the measles of traditional Chinese medicine[D].Beijing:Beijing University of Technology, 2013.

[7]周志华.机器学习[M].北京：清华大学出版社，2016.

[8]张艺凡,胡广芹,张新峰.基于支持向量机的痤疮患者舌色苔色识别算法研究[J].北京生物医学工程,2016,35(1):7-11.

Zhang YF, Hu GQ , Zhang XF . An algorithm study on tongue color recognition of patients with acne based on support vector machine[J]. Beijing Biomedical Engineering, 2016,35(1):7-11.

[9]李颀,王俊.基于颜色和纹理特征融合的马铃薯分级方法[J].科学技术与工程,2019,19(25):273-279.

Li Q ,Wang J . Potato classification method based on fusion of color and texture features[J]. Science Technology and Engineering,2019,19(25):273-279.

[10]?Markus A. Stricker, markus orengo. similarity of color images[J]. Proc Spie Storage & Retrieval for Image & Video Databases, 1995, 2420:381--392.

[11]?Zhan H,Luo ZT,He ZS ,et al. Tamura coarseness for evaluating OTH radar image and its application in RFI suppression[C]. 2019 IEEE Radar Conference ,2019:1-6.

[12]?罗小燕,陈慧明,卢小江,等.基于网格搜索与交叉验证的SVM磨机负荷预测[J].中国测试,2017,43(1):132-135，144.

Luo XY, Chen HM , Lu XJ , et al. Forecast of SVM mill load based on grid search and cross validation[J]. China Measurement & Test, 2017,43(1):132-135，144.

[13]Krizhevsky?A, Sutskever?I, Hinton?GE. ImageNet classification with deep convolutional neural networks[C]//?Pereira?F，Burges?CJC，Bottou?L，et al.?Proceedings of the 25th International Conference on Neural Information Processing Systems.?Red Hook，NY：Curran Associates Inc，?2012:1097-1105.

[14]Zhang X,Pan W,?Xiao P. In-vivo skin capacitive image classification using AlexNet convolution neural network[C].?2018 IEEE 3rd International Conference on Image, Vision and Computing.?Chongqing：IEEE?Xplore, Ei and Scopus，2018:439-443.

[15]侯霄雄, 许新征, 朱炯,等. 基于AlexNet和集成分类器的乳腺癌计算机辅助诊断方法[J]. 山东大学学报（工学版）, 2019, 49(2):21-25.

Hou XX?, Xu XZ?, Zhu J , et al. Computer aided diagnosis method for breast cancer based on AlexNet and ensemble classifiers[J]. Journal of Shandong University(Engineering Science),2019, 49(2):21-25.

[16]罗金梅,罗建,李艳梅,等.基于多特征融合CNN的人脸识别算法研究[J].航空计算技术,2019,49(3):40-45.

Luo?JM, Luo?J, Li?YM. Faceｒecognition algorithm based on multi- feature fusion convolution neural network[J].Aeronautical Computing Technique,2019,49(3):40-45.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录