Heart Energy Signature is very Different from a Standard Spectrogram
Typical clinical complaint with the spectrogram is that it can not separate S1 from the
murmur or allow to separate S2 clearly or to see A2 and P2 components clearly. Same
goes for S3 and S4 sounds. We show example spectrogram below to illustrate that.
S1 Murmur S2 S3
Original recording (using third party Music Editing Software)
Waveletogram (obtained using wavelet transform)
Spectrograms display frequency as vertical axis and time as horizontal and plot sound intensity in decibel in color. They utilize Short Term Window Fourier Transform (SWFT) that is a first order
method and is subject to the uncertainty principle, whereby one can not achieve simultaneous resolution in frequency and in time. Correspondingly, spectrogram image is window size
dependent. See images below (as window size increases time resolution decreases and
frequency resolution improves).
Window size: 1024
Window Size: 2048
Window Size: 4096
Original Heart Sound Recording
Heart Energy Signature Method Spectrogram of the same heart sound recording is shown below.
It is a second order method and can simultaneously resolve both time and frequency and can
separate all key components of the heart sound as it is shown on the figures below.
S1 Murmur S2(sp) S3 S1 murmur S2
Murmur Split S2 S3
Further detail shows first heart beat without S1 - murmur, S2 and S3. Note clear separation of Murmur,
A2 and P2 components. S2 and S3 are clearly visible.
Also S1 shows clear separation of mitral and tricuspid components on the Energy
Power vs. Time Plot
S1: (1) (2)
and S2 also shows clear A2 and P2 on the same plot
A2 P2 S3