Study of Feature Extraction Algorithms on Photoplethysmography (PPG) Signals to Detect Coronary Heart Disease

Abstract

Coronary Heart Disease (CHD) is the most dangerous heart disease, this disease occurs, when the blood supply containing oxygen and nutrients to the heart muscle blocked by plaque in the heart blood vessels or coronary arteries. Currently, there are many ways of diagnosing coronary heart disease, starting from using ECG to Cardiac catheterization. However, it has some drawbacks, including the inflexibility of diagnosing quickly and invasive procedures. Heart rate variability (HRV) is a strong indication of cardiovascular diseases; as a result, any change in the normal heart rate (or blood volume) activity is a major marker for a potential cardiovascular malfunction. Through a series of waves and peak detection, photoplethysmography (PPG) detects blood pressure, oxygen saturation, and cardiac output. In recent years, there have been more studies using ECG signals to detect CHD compared to PPG signals, especially those discussing feature extraction on PPG signals in detecting CHD because this greatly affects the accuracy of CHD detection. In this study, proposed a literature study of feature extraction algorithm for detecting coronary heart disease using photoplethysmography. For the feature extraction, three algorithm will be discussed are respiratory rate (RR) interval, HRV Features and Time Domain Features. HRV features, with 94.4% accuracy, 100% sensitivity, and 90.9% specificity, is the best feature extraction approach of the three proposed techniques using decision tree classifier.