Defective Heartbeat Detection

Abstract

In this project, the aim is to detect the defective heartbeat by a non-invasive method. Preliminary detection of the heartbeat can help in various stages afterwards and can lead to high performance heart diagnostics. The Fast-Fourier transform of a defective ECG is found out and is compared against the FFT of a normal ECG. Time-domain analysis is observed by calculating the power spectral density after noise is removed from the ECG and then it is compared against a normal ECG with noise removed as well. Similarly, Auto-correlation of both defective and normal ECG are also found out and then their crosscorrelation helps in establishing relationships between the two wave-forms. All the above mentioned parameters helps in establishing relationships between the defective and normal ECG. And such procedures with pre-processing techniques implemented can help in developing of a non-invasive tool for detecting a defective heart-beat.