Detection of coronary occlusions using autoregressive modeling of diastolic heart sounds

Metin Akay,J. Semmlow,W. Welkowitz,M. Bauer,J. B. Kostis
DOI: https://doi.org/10.1109/10.52343
IF: 4.6
1990-04-01
IEEE Transactions on Biomedical Engineering
Abstract:Recordings of diastolic heart sound segments were modeled by autoregressive (AR) methods including the adaptive recursive least-squares lattice (RLSL) and the gradient lattice predictor (GAL). Application of the Akaike criterion demonstrated that between 5 and 15 AR coefficients are required to describe a diastolic segment completely. The reflection coefficients, prediction coefficients, zeros of the polynomial of the inverse filter, and AR spectrum were determined over a number (N=20-30) of diastolic segments. Preliminary results indicate that the averaged AR spectrum and the zeros of the inverse filter polynomial can be used to distinguish between normal patients and those with coronary artery disease.<<ETX>>
What problem does this paper attempt to address?