[1]张强,魏高峰△,闫士举△,等.基于机器学习方法的超声M模式气胸图像的分类研究[J].生物医学工程研究,2022,02:151-157.
ZHANG Qiang,WEI Gaofeng,YAN Shiju,et al.Classification of ultrasound M-mode pneumothorax images based on machine learning methods[J].Journal of Biomedical Engineering Research,2022,02:151-157.
点击复制
基于机器学习方法的超声M模式气胸图像的分类研究(PDF)
《生物医学工程研究》[ISSN:1006-6977/CN:61-1281/TN]
- 期数:
-
2022年02期
- 页码:
-
151-157
- 栏目:
-
- 出版日期:
-
2022-06-25
文章信息/Info
- Title:
-
Classification of ultrasound M-mode pneumothorax images based on machine learning methods
- 文章编号:
-
1672-6278 (2022)02-0151-07
- 作者:
-
张强1; 魏高峰2△; 闫士举1△; 张涛1; 汪俊豪1
-
1.上海理工大学健康科学与工程学院,上海 200093;2.海军军医大学海军医学系,上海 200433
- Author(s):
-
ZHANG Qiang1 ; WEI Gaofeng2 ; YAN Shiju1 ; ZHANG Tao1 ; WANG Junhao1
-
1.School of Health Science and Engineering,University of Shanghai for Science and Technology, Shanghai 200093,China;2.Department of Naval Medicine, Naval Medical University, Shanghai 200433
-
- 关键词:
-
M超声图像; 特征提取; 传统分类算法; 气胸诊断; 分类器; 沙滩征
- Keywords:
-
M-ultrasound image; Feature extraction; Traditional classification algorithm; Pneumothorax diagnosis; Classifier; Beach sign
- 分类号:
-
R318;TP301
- DOI:
-
10.19529/j.cnki.1672-6278.2022.02.08
- 文献标识码:
-
A
- 摘要:
-
本研究基于图像提取特征结合机器学习方法,建立超声M模式图像分类模型,为气胸诊断提供参考。收集肺部滑动存在特征典型图像171幅,特征不典型图像283幅;肺部滑动消失特征典型图像1 113幅,特征不典型图像111幅;肺点特征典型图像850幅,特征不典型图像285幅。通过提取灰度共生矩阵、灰度游程矩阵等纹理特征,采用五折交叉验证方法,使用随机森林、朴素贝叶斯和支持向量机3种分类器对M模式下超声图像进行分类。在使用支持向量机下,对单独特征典型图像进行分类的准确率最高,达到99.2%,灵敏度为99.54%,特异性为97.08%。实验结果证明,机器学习有望作为一种新的辅助诊断手段,有助于提高急救场合下的超声诊断气胸的准确率。
- Abstract:
-
Based on image extraction features combined with machine learning methods, We established an ultrasound M-mode image classification model to provide a reference for pneumothorax diagnosis. We collected 171 images with typical features of lung sliding, 283 images with atypical features; 1 113 images with typical features of lung sliding disappearing, 111 images with atypical features; 850 images with typical features of lung point, 285 images with atypical features. We classified ultrasond images in M-mode by extracting texture features such as gray-level co-occurrence matrix and gray-level run-length matrix, using method of five-fold cross validation and three kinds of classifiers-random forest, naive bayes and support vector machine. Under the use of SVM, the classification of images of typical features had the highest accuracy of up to 99.2%, with a sensitivity of 99.54% and a specificity of 97.08%. The results of experiment show that machine learning is expected to be used as a new auxiliary diagnosis method to help improve the ultrasound diagnosis accuracy of pneumothorax in emergency.
备注/Memo
- 备注/Memo:
-
(收稿日期:2022-01-20)△通信作者Email: yanshj99@aliyun.com; highpeak8848@163.com
更新日期/Last Update:
2022-07-21