Ashok K. Singh (Hospitality) and Rohan J. Dalpatadu (Math) published their study, in Machine Learning with Applications. The study addresses the problem of predicting the risk of obstructive sleep apnea from overnight breath recordings collected by the smartphone app ZeeAppnea. Four data mining multi-level classifiers were used on the Fast Fourier Transform (FFT) of each time series, and prediction accuracies were computed. The results showed that classifiers Random Forest (RF) or Support Vector Machine (SVM) can be used on the recordings obtained from ZeeAppnea instead of the time-consuming manual interpretation of charts of breathing amplitudes by medical personnel, as this would improve prediction accuracy and automate the process of this screening application.
