Premium
Abnormal pattern detection in PPG signals during sports and exercises
Author(s) -
Quan Long,
Wu Yunming
Publication year - 2020
Publication title -
internet technology letters
Language(s) - English
Resource type - Journals
ISSN - 2476-1508
DOI - 10.1002/itl2.240
Subject(s) - photoplethysmogram , hypovolemia , signal (programming language) , breathing , heart rate , pattern recognition (psychology) , pulse rate , character (mathematics) , pulse (music) , artificial intelligence , computer science , medicine , biomedical engineering , speech recognition , mathematics , computer vision , blood pressure , anatomy , telecommunications , geometry , filter (signal processing) , detector , programming language
Pulse oximeter is an important clinical equipment, which can measure the photoplethysmogram (PPG) signal to calculate the heart rate, blood volume, breathing, hypovolemia etc. However, the PPG signal is easily interfered by some factors, such as body movement in sports or exercises. Before calculating the heart rate, blood volume, breathing, or hypovolemia, we need to recognize these interfered segments. In order to identify these abnormal patterns, the PPG signal is converted as a symbol sequence which consists of several characters. Then, the segment of PPG signal is represented as a character sequence. The abnormal pattern in PPG signal is detected by the distance between character sequences which are obtained from local and global information. When the distance is great than a threshold, the segment is regarded as an abnormal pattern. The proposed method also can be used in other physiological signals.