| 许 维,吕倩倩,江钟立,林 枫.基于声学分析和机器学习构建咳嗽和清嗓分类模型[J].中国康复医学杂志,2020,(12):1434~1438 |
| 基于声学分析和机器学习构建咳嗽和清嗓分类模型 点此下载全文 |
| 许 维 吕倩倩 江钟立 林 枫 |
| 南京医科大学附属逸夫医院，江苏省南京市，211100 |
| 基金项目：南京医科大学科技发展基金（NMUB2018295）；江苏省高校哲学社会科学优秀创新团队建设项目（2017STD006） |
| DOI：10.3969/j.issn.1001-1242.2020.12.005 |
| 摘要点击次数: 948 |
| 全文下载次数: 630 |
| 摘要： |
| 摘要
目的：建立气道廓清动作的声学分类器，为实施肺康复的咳嗽训练提供监测工具。
方法：健康男性11例和女性15例，分别在平卧位、45°靠坐位和90°端坐位，根据随机视觉指令执行咳嗽和清嗓动作各10次，并同时录制声音，分析声音片段的时域、频域和信息域特征，由此构建声音的特征矢量用于机器学习。采用的模型包括：线性判别分析、分类回归决策树、随机森林和线性分类器型支持向量机。
结果：模型间比较显示随机森林方法所建分类器具有更高的准确度（0.9162）和一致性（Kappa值为0.8323）。验证结果显示该模型无论在区分体位因素或不区分体位情况下，对咳嗽音有较高的准确度、一致性、敏感度和特异度。
结论：咳嗽和清嗓动作具有声学差异，并且这种差异可以由随机森林方法构建机器学习模型加以分类，由此为肺康复治疗中采用声学手段辅助判断气道廓清动作类型提供了依据。 |
| 关键词：咳嗽音 清嗓音 机器学习 声学分析 |
| Classifying cough and throat clearing behaviors based on acoustic analysis and machine learning Download Fulltext |
|
| Sir Run Run Hospital， Nanjing Medical University， Jiangsu， Nanjing， 211100 |
| Fund Project: |
| Abstract: |
| Abstract
Objective: To establish an acoustic classifier of airway clearance related behaviors and to provide a monitoring tool for cough training in pulmonary rehabilitation.
Method: Eleven healthy males and fifteen females were seated in flat，45-degree and 90-degree positions， respectively. Cough and throat clearing behaviors were performed 10 times each according to the random visual instruction， and the sound was recorded at the same time. For each sound segment， acoustic features of time-domain， frequency-domain and information-domain were analyzed， then extracted the feature vectors of sound for machine learning. The models contain linear discriminant analysis， classification regression decision tree， random forest and linear classifier type support vector machine.
Result: The comparison between the models showed that the classifiers established by random forest method had higher accuracy (0.9162) and the Kappa value (0.8323). Whether or not the influences of postural factors are included， the verification results showed that random forest model had the highest accuracy， the Kappa value， sensitivity and specificity for classification.
Conclusion: There are acoustic differences between cough and throat clearing， and the differences can be classified by machine learning model. Our study provides a promising strategy for developing airway clearance behavior related classifier based on acoustic information. |
| Keywords:cough sound throat clearing sound machine learning acoustic analysis |
|
| 查看全文 查看/发表评论 |
