Heart Disease Detection from Human Voice

Atrial fibrillation is the most common form of abnormal heart rhythm. The disease affect the atrial in the heart by making their beat faster and disordered, which in results affect the blood flow and create the distorted heartbeat. The disease itself isn't fatal, however in the worst case scenario it can cause stroke or even death. AF does not have visible symptoms and can only be diagnosed by invasive methods, which includes ECG and constant tracking of the heartbeat.
The goal of the project was to examine an alternative diagnosis which is based on the patient speech sample, to be exact, based on the excitation signal of the speech – the Glottal Pulse. The reason for this method is because the distortion only happens for a short time every once in a while, so there is a need for a quick and easier diagnostic The project include two parts, the first is the extraction of the Glottal Pulse by estimating an LTI system and use inverse filtering on the speech signal. The second part is analyzing the Glottal Pulse by calculating different parameters of the signal, and using the said parameters to and classify speech samples between healthy and sick patients. In addition, there will be an analysis to check if the Heartbeat affect the Glottal Pulse.

Heart Disease Detection from Human Voice
Heart Disease Detection from Human Voice
Heart Disease Detection from Human Voice