A Novel Detection Approach for Cardio-Respiratory Disorders Using PPG Signals
The aim of the study was to determine the use of Photoplethysmography (PPG) as a tool for identifying cardiac and respiratory disorders using Decision tree mining technique. PPG signals were recorded from 45 healthy volunteers in the 19-22 age group. Recordings were carried out under normal, induced cardiac stress and induced apnea conditions to assess the changes in the PPG morphology under these settings. Three features, stiffness index (SI), reflection index (RI) and power ratio (PR), have been used for classification. Classification accuracy of 94.44% and 97.19% has been achieved for induced cardiac stress and induced apnea recordings respectively, using the decision tree classifier. The study indicates that PPG can be used as an effective screening tool for preliminary diagnosis of different cardiac and respiratory conditions. The results need to be validated for large datasets as well as for offline analysis of measurements from real life situations.