TANG Yuqi,LI Zechen,YANG Dongdong,et al.Research and Application of Facial Motor Nerve Conduction Examination Data based on Machine Learning[J].Journal of Chengdu University of Information Technology,2020,35(05):519-523.[doi:10.16836/j.cnki.jcuit.2020.05.007]
基于机器学习的面部运动神经传导检查数据的研究及应用
- Title:
- Research and Application of Facial Motor Nerve Conduction Examination Data based on Machine Learning
- 文章编号:
- 2096-1618(2020)05-0519-05
- Keywords:
- machine learning; facial nerve; electromyography; characteristic; random forests
- 分类号:
- TP181
- 文献标志码:
- A
- 摘要:
- 为初步研究面部运动神经传导检查数据,提出运用机器学习方法进行深度数据挖掘、分析,找出相关性最高的特征值,以研究其主要的影响因素及探讨临床诊断预测的可能性。收集成都中医药大学附属医院10个月的肌电检查报告共2352份数据,筛选符合标准的 575 份报告,制作数据集,利用编程的方式对其检查数据和报告结论进行量化分析,分别建立KNN、逻辑回归、随机森林、stacking 算法模型,经过调参选取正确率最高的模型进行特征提取以研究其主要影响因素及研究临床判读预测的可能性。实验结果表明,一方面在肌电图临床判读中随机森林算法正确率达到92.69%,精度为92.78%,召回率为100%,与逻辑回归相比较P值为0.04271,与KNN相比较P值为0.00745,均具有显著统计学意义,即随机森林模型最适合于面部运动传导神经检查数据分析。另一方面,运用随机森林方法提取特征值,能够更加清晰迅速地找出影响面部运动神经病变的最主要因素。通过机器学习挖掘数据,得出影响面部运动神经传导检查的主要因素为颞支右侧波幅数据和颊支右侧波幅等8个特征点,并提出可使用临床获得的数据集进行判读预测并通过随机森林选取主要的特征点,具体以减少临床操作时检查点位的形式达到缩短单人检查时间的目的。
- Abstract:
- In order to conduct a preliminary study on facial motor nerve conduction examination data,the method of machine learning is proposed to conduct in-depth data mining and analysis to find the characteristic values with the highest correlation, so as to study the main influencefactors and explore the possibility of clinical diagnosis prediction. We collected 10 months electrical inspection report totally 2352 datafromchengdu university of traditional Chinese medicine medical and selected 575 reports accord with standard to make datasets,we use programming to quantitatively analyze the inspection data and report conclusion,and building KNN,logistic regression,random forests,stacking algorithm model respectively.The model with highest accuracy is selected to extract features after adjustment of arguments, toresearch the main influence factors and the clinical interpretation prediction possibilities.Experimental results show that,on the one hand,random forest algorithm in emg clinical interpretation accuracy reached 92.69%,the precision is 92.78%,the recall rate was 100%,P value is 0.04271 compared with logistical regression and 0.00745 compared with KNN,mentioned above all make great sense.,namely the random forest model is best suited for facial motor conduction nerve examination data analysis,on the other hand,the use of random forests method to extract the characteristic value,can more clear quickly find out the main factors affecting facial movement neuropathy.This research mines data through machine learning,it is concluded that the main factors affecting facial motor nerve conduction check are temporal right amplitude data and buccal branch on the right side of amplitude and so on eight feature points.It is proposed that the clinical data sets can be used for interpretation prediction.By randomly selecting main feature points,specifically reduce the inspection points in clinical operations,to achieve the goal of reducing single person inspection time.
参考文献/References:
[1] 严律南.肝脏外科发展三阶段论及中国医生的贡献[J].中国普外基础与临床杂志,2018,25(4):388-390.
[2] De Silva K,Jönsson D,Demmer RT.A combined strategy of feature selection and machine learning to identify predictors of prediabetes[J].J Am Med Inform Assoc,2020,27(3):396-406.
[3] Dinh A,Miertschin S,Young A,et al.A data-driven approach to predicting diabetes and cardiovascular disease with machine learning[J].BMC Med Inform Decis Mak,2019,19(1):211.
[4] van Os HJA,Ramos LA,Hilbert A,et al.Predicting Outcome of Endovascular Treatment for Acute Ischemic Stroke:Potential Value of Machine Learning Algorithms[J].Front Neurol,2018(9):784.
[5] 何易晨,黄杰,熊仕奇,等.慢性阻塞性肺疾病的健康管理系统设计[J].成都信息工程大学学报,2018,33(3):250-257.
[6] Kononenko I.Machine learning for medical diagnosis:history,state of the art and perspective[J].Artif Intell Med,2001(23):89-109.
[7] Powell S,Magnotta VA,Johnson H,et al.Registration and machine learning-based automated segmentation of subcortical and cerebellar brain structures[J].Neuroimage,2008(39):238-247.
[8] Yaman E,Subasi A.Comparison of Bagging and Boosting Ensemble Machine Learning Methods for Automated EMG Signal Classification[J].Biomed Res Int,2019:13.
[9] 李玲,凡子莲,戴东璟,等.听性脑干反应V波反应阈应用于听力测定的意义[C].国际数字医学会、Digital Chinese Medicine.湖南中医药大学学报2016/专集:国际数字医学会数字中医药分会成立大会暨首届数字中医药学术交流会论文集.国际数字医学会、Digital Chinese Medicine:湖南中医药大学期刊杂志社,2016:666-667.
[10] 文宏伟,陆菁菁,何晖光.机器学习在神经精神疾病诊断及预测中的应用[J].协和医学杂志,2018,9(1):19-24.
[11] 党静霞.肌电图诊断与临床应用[M].北京:人民卫生出版社,2005(12):17-484.
[12] 王心刚.神经电生理诊断技术规范[J].现代电生理学杂志,2017,24(2):117-126.
[13] 王心刚.神经电生理诊断技术规范(续)[J].现代电生理学杂志,2018,25(1):55-62.
[14] 杨丽艳,秦少福,罗苑媚.肌电图对面神经麻痹患者病情评估的研究进展[J].神经病学与神经康复学杂志,2013,10(1):31-32.
[15] 叶爱萍.面神经麻痹肌电图及面神经传导速度的观察[J].中国临床康复,2003(1):106.
[16] 林恬恬.数据挖掘定量与定性相结合对李平教授冠心病诊疗规律的研究[D].北京:北京中医药大学,2018.
[17] Song Y,Huang J,Zhou D,et al.IKNN:Informative K-Nearest Neighbor Pattern Classification[C].Knowledge Discovery in Databases:PKDD 2007,11th European Conference on Principles and Practice of Knowledge Discovery in Databases,Warsaw,Poland,September 17-21,2007,Proceedings.DBLP,2007.
[18] 戴维.逻辑回归解决文本分类问题[J].通讯世界,2018(8):266-267.
[19] 瞿海斌,毛利锋,王阶.基于决策树的血瘀证诊断规则自动归纳方法[J].中国生物医学工程学报,2005(6):709-711.
相似文献/References:
[1]李宝林,周 坤,李仕伟.一种基于M-Bisearch的最大频繁项集挖掘算法研究[J].成都信息工程大学学报,2016,(05):463.
LI Bao-lin,ZHOU Kun,LI Shi-wei.Research on Mining Algorithm of Maximal Frequent Itemsets based on M-blsearch[J].Journal of Chengdu University of Information Technology,2016,(05):463.
[2]毛开银,赵长名,何 嘉.基于XGBoost的10 m风速订正研究[J].成都信息工程大学学报,2020,35(06):604.[doi:10.16836/j.cnki.jcuit.2020.06.004]
MAO Kaiyin,ZHAO Changming,HE Jia.A Research for 10 m Wind Speed Prediction based on XGBoost[J].Journal of Chengdu University of Information Technology,2020,35(05):604.[doi:10.16836/j.cnki.jcuit.2020.06.004]
[3]李孝涌,陈科艺,李熙晨.卷积神经网络在ENSO预报中的应用[J].成都信息工程大学学报,2022,37(01):81.[doi:10.16836/j.cnki.jcuit.2022.01.014]
LI Xiaoyong,CHEN Keyi,LI Xichen.Application of Convolutional Neural Network in ENSO Prediction[J].Journal of Chengdu University of Information Technology,2022,37(05):81.[doi:10.16836/j.cnki.jcuit.2022.01.014]
备注/Memo
收稿日期:2020-03-23
基金项目:四川省教育厅重大培育资助项目(18CZ0021)