Quality-driven wavelet based PCG signal coding for wireless cardiac patient monitoring

In this paper, we present a quality driven PCG signal coding scheme for wireless cardiac patient monitoring applications. The proposed quality driven codec is designed based on the wavelet-based compression method and the wavelet energy based diagnostic distortion (WEDD) measurement criterion. The proposed WEDD measure is the weighted percentage root mean square difference between the wavelet subband coefficients of the original and compressed signals with weights equal to the relative wavelet subband energies of the corresponding subbands. The WEDD measure appears to be a correct representation of the amount of signal distortion at all the subbands, and robust to insignificant errors in some bands. The performance of the proposed method is validated using the PCG signal blocks taken from the qdheart database and CAHM database PCG records which include many different valvular pathologies such as normal sounds, late systolic, ejection click, tricuspid regurgitation, diastolic aortic insufficiency, murmurs, and noises. Results showed that the performance of the WEDD criterion outperforms the PRDw and WWPRD criteria. For WEDD=4%, the maximum compression ratio of 186.07 was achieved for the test signal from the Diastolic Fixed S2 Split II record and the minimum compression ratio of 21.16 is obtained for the signal from the Diastolic Atrial Septal Defect record.