期刊
SENSORS
卷 22, 期 13, 页码 -出版社
MDPI
DOI: 10.3390/s22134919
关键词
automated machine learning; hyperparameter optimization; electrocardiogram; time series; ST-segment; myocardial infarction
资金
- ITEA INNO4HEALTH Project [19008]
- SURF Cooperative [EINF-2491]
- TAILOR - EU Horizon 2020 research and innovation programme [952215]
In this study, the focus is on ST-segment deviation as an indicator of myocardial infarction, proposing a methodology to detect and quantify this abnormality through machine learning. Validation on the ST-T database from Physionet showed high accuracy rates, making the method promising for further applications.
Nowadays, even with all the tremendous advances in medicine and health protocols, cardiovascular diseases (CVD) continue to be one of the major causes of death. In the present work, we focus on a specific abnormality: ST-segment deviation, which occurs regularly in high-performance athletes and elderly people, serving as a myocardial infarction (MI) indicator. It is usually diagnosed manually by experts, through visual interpretation of the printed electrocardiography (ECG) signal. We propose a methodology to detect ST-segment deviation and quantify its scale up to 1 mV by extracting statistical, point-to-point beat characteristics and signal quality indexes (SQIs) from single-lead ECG. We do so by applying automated machine learning methods to find the best hyperparameter configuration for classification and regression models. For validation of our method, we use the ST-T database from Physionet; the results show that our method obtains 98.30% accuracy in the case of a multiclass problem and 99.87% accuracy in the case of binarization.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据