Segmentation of heart sound recordings by a duration-dependent hidden Markov model

TLDR

Digital stethoscopes enable computerized heart‑sound analysis, yet handheld recordings in clinical settings are noisy, complicating segmentation of S1 and S2. The study proposes a duration‑dependent hidden Markov model to robustly segment heart sounds. The DHMM infers the most likely sequence of physiological heart sounds by combining event duration, signal‑envelope amplitude, and a predefined model structure, and was trained and tested on bedside recordings from patients undergoing coronary angiography. In 73 patients, the DHMM achieved 98.8 % sensitivity and 98.6 % positive predictive value, correctly identifying 890 of 901 S1/S2 sounds and misplacing only 13 of 903, confirming its suitability for clinical heart‑sound segmentation.

Abstract

Digital stethoscopes offer new opportunities for computerized analysis of heart sounds. Segmentation of heart sound recordings into periods related to the first and second heart sound (S1 and S2) is fundamental in the analysis process. However, segmentation of heart sounds recorded with handheld stethoscopes in clinical environments is often complicated by background noise. A duration-dependent hidden Markov model (DHMM) is proposed for robust segmentation of heart sounds. The DHMM identifies the most likely sequence of physiological heart sounds, based on duration of the events, the amplitude of the signal envelope and a predefined model structure. The DHMM model was developed and tested with heart sounds recorded bedside with a commercially available handheld stethoscope from a population of patients referred for coronary arterioangiography. The DHMM identified 890 S1 and S2 sounds out of 901 which corresponds to 98.8% (CI: 97.8–99.3%) sensitivity in 73 test patients and 13 misplaced sounds out of 903 identified sounds which corresponds to 98.6% (CI: 97.6–99.1%) positive predictivity. These results indicate that the DHMM is an appropriate model of the heart cycle and suitable for segmentation of clinically recorded heart sounds.