Predicting Pulmonary Function From the Analysis of Voice: A Machine Learning Approach.
Predicting Pulmonary Function From the Analysis of Voice: A Machine Learning Approach.
复制标题
DOI：
10.3389/fdgth.2022.750226
复制
发表时间：
2022
影响因子：
--
通讯作者：
Rezwan FI
中科院分区：
文献类型：
--
作者：
Alam MZ;Simonetti A;Brillantino R;Tayler N;Grainge C;Siribaddana P;Nouraei SAR;Batchelor J;Rahman MS;Mancuzo EV;Holloway JW;Holloway JA;Rezwan FI
为了自我监测哮喘症状，现有方法（例如峰值流量计、智能肺活量计）需要特殊设备，并且患者并不总是使用。语音记录有可能产生肺功能的替代指标，本研究旨在应用机器学习方法来预测哮喘患者的肺功能和肺功能异常的严重程度。设计了一种基于阈值的机制来从323个录音中分离语音和呼吸。从中提取的特征与生物学因素相结合，以预测肺功能。使用随机森林（RF）、支持向量机（SVM）和线性回归算法开发了三种预测模型：（a）预测肺功能的回归模型，（B）预测肺功能异常严重程度的多类分类模型，和（c）预测肺功能异常的二元分类模型。将训练样本和测试样本分开（70%：30%，使用平衡分配），将特征归一化，使用10倍交叉验证，并在测试样本上评估模型性能。基于RF的回归模型表现更好，最低均方根误差为10·86。为了预测肺功能损害的严重程度，基于SVM的模型在多类分类中表现最好（准确度= 73.20%），而基于RF的模型在预测异常肺功能的二进制分类模型中表现最好（准确度= 85%）。我们的机器学习方法可以从录制的语音文件中预测肺功能，比已发表的方法更好。这项技术可用于开发未来的远程医疗解决方案，包括基于智能手机的应用程序，这些应用程序有可能帮助哮喘患者做出决策和自我监测。
To self-monitor asthma symptoms, existing methods (e.g. peak flow metre, smart spirometer) require special equipment and are not always used by the patients. Voice recording has the potential to generate surrogate measures of lung function and this study aims to apply machine learning approaches to predict lung function and severity of abnormal lung function from recorded voice for asthma patients. A threshold-based mechanism was designed to separate speech and breathing from 323 recordings. Features extracted from these were combined with biological factors to predict lung function. Three predictive models were developed using Random Forest (RF), Support Vector Machine (SVM), and linear regression algorithms: (a) regression models to predict lung function, (b) multi-class classification models to predict severity of lung function abnormality, and (c) binary classification models to predict lung function abnormality. Training and test samples were separated (70%:30%, using balanced portioning), features were normalised, 10-fold cross-validation was used and model performances were evaluated on the test samples. The RF-based regression model performed better with the lowest root mean square error of 10·86. To predict severity of lung function impairment, the SVM-based model performed best in multi-class classification (accuracy = 73.20%), whereas the RF-based model performed best in binary classification models for predicting abnormal lung function (accuracy = 85%). Our machine learning approaches can predict lung function, from recorded voice files, better than published approaches. This technique could be used to develop future telehealth solutions including smartphone-based applications which have potential to aid decision making and self-monitoring in asthma.
登录
查看更多内容
DOI：
10.1109/ojemb.2020.3026928
发表时间：
2020
影响因子：
5.8
作者：
Laguarta J;Hueto F;Subirana B
通讯作者：
Subirana B
影响因子：
3.2
作者：
Sharan, Roneel V.;Abeyratne, Udantha R.;Porter, Paul
通讯作者：
Porter, Paul
影响因子：
3.1
作者：
Tayler, Nicholas;Grainge, Christopher;Holloway, Judith
通讯作者：
Holloway, Judith
影响因子：
1.5
作者：
Wiechern, Beth;Liberty, Kathleen A.;Lin, Emily
通讯作者：
Lin, Emily
影响因子：
--
作者：
BELL, D;LAYTON, AJ;GABBAY, J
通讯作者：
GABBAY, J