Premium
ECG signal analysis for fatigue and abnormal event detection during sport and exercise
Author(s) -
Luo Xi
Publication year - 2020
Publication title -
internet technology letters
Language(s) - English
Resource type - Journals
ISSN - 2476-1508
DOI - 10.1002/itl2.262
Subject(s) - support vector machine , signal (programming language) , wearable computer , event (particle physics) , computer science , artificial intelligence , frequency domain , state (computer science) , speech recognition , set (abstract data type) , time domain , pattern recognition (psychology) , computer vision , algorithm , embedded system , physics , quantum mechanics , programming language
With the continuous improved quality of life, more and more persons have paid attention on their health. Sport and exercise have become a popular way to keep health. However, improper and excessive sport and exercise may cause physical injury. In order to solve this issue, this paper proposes a system to monitor the state during sport and exercise for detecting fatigue and abnormal event. If a person is under fatigue state, he or she stops exercising to rest. In the fatigue detection system, first the electrocardiograph (ECG) signal is collected by a smart wearable device, then the noises in the signal are removed by several filters and the denoised signal are represented as nine features from time domain and frequency domain, lastly the extracted features are input into a trained weighted one‐class support vector machine (WOC‐SVM) model to determine the state during sport. The WOC‐SVM is learnt offline on a training set which consists of many collected data. The experimental results show the effectiveness of the proposed system.