Abstract:
The most dangerous of heart diseases ventricular tachyarrhythmia (VT) is very difficult
to detect. Our project, an amalgamation of research and application development uses a
novel wavelet based algorithm for detecting VT. This project is done in collaboration
with Armed Forces Institute of Cardiology, National Institute of Heart Diseases (AFICNIHD)
Wavelet transform has emerged over recent years as a powerful time–frequency analysis
tool favored for the interrogation of complex non-stationary signals. The proposed
algorithm uses an efficient method for detecting VT in wavelet pre-processed ECG
signals. A MATLAB routine using built in library functions for pre-processing removes
high frequency noise. The preprocessed signal is applied to the Spectral Algorithm
(SPEC) which works in frequency domain and analyses the energy content. If the
algorithm decides that the ECG part contains VT, the result is accepted as true and no
further investigation is required. Otherwise a further investigation is carried out to
confirm the result or disprove it. The terminal parts of the ECG signal are processed with
a continuous wavelet transform, which leads to a time-frequency representation of the
signal. The diagnostic feature vectors are obtained by subdividing the representations into
several regions and by processing the sum of the decomposition coefficients belonging to
each region. Wavelet based efficient algorithms are used for detection of VT. With these
methods, underlying features within the VT waveform are made visible in the wavelet
time-scale half space. The proposed algorithms overcome the non-sensitivity of SPEC
iii
algorithm utilizing its highly specific nature to the fullest. An exhaustive testing
exemplified higher sensitivity, predictivity and specificity, enabling the cardiologists and
electro physiologists to detect VT with accuracy of more than 85%.
