Key result
An unsupervised segmentation method based on abrupt changes detection achieved mean F1 scores of 91.2%, 94.3%, and 96.3% across three databases of ~3,500 heart sound signals.
Why the study?
Automated segmentation of cardiograms into fundamental heart states can guide physicians to analyze a patient's condition more effectively, but simple and robust unsupervised methods are needed.
A novel unsupervised algorithm for segmenting heart sounds demonstrated high accuracy across multiple databases, potentially aiding automated auscultation analysis.
May aid automated heart sound analysis for early diagnosis; leaves open prospective clinical validation before practice change.
Heart auscultation enables early diagnosis of cardiovascular diseases. Automated segmentation of cardiograms into fundamental heart states can guide physicians to analyze the patient’s condition more effectively. In this work, we propose an unsupervised method of segmentation into heart sounds and silences based on the detection of abrupt changes in the signal. Our procedure involves two steps. First, the abrupt changes, which correspond to the beginning and end of the heart sounds, are localized. Heart sounds and silences are then identified by calculating the signal power in each interval defined by the change points. The parameters of our algorithm are adjusted on the basis of estimated heart rate alone. We evaluate our method on three independent open-access database (PhysioNet 2016, CirCor DigiScope and PASCAL) for healthy and pathological populations, with or without murmurs. It achieves mean F 1 score detection performance of 91.2%, 94.3% and 96.3% respectively, outperforming most of the competing unsupervised approaches. By providing top ranking detection performance for three different types of heart sounds database, the proposed algorithm is reliable and robust, yet easy to implement. This paper presents a simple and effective alternative segmentation method that can help improve the physiological interpretation of heart sounds recordings. • A novel unsupervised approach is proposed for the segmentation of cardiogram signals. • Its principle is based on the detection of abrupt changes in the recordings. • The parameters of the algorithm can be set on the basis of estimated heart rate alone. • The method is evaluated on healthy and pathological subjects, with or without murmurs. • It delivers top-rank performance for three databases including ∼ 3,500 signals.
No takes yet. Share an insight, caveat, or question.
Tatulli et al. (2025) studied Heart sounds (healthy and pathological, with or without murmurs) (n=3,500). Unsupervised segmentation method based on abrupt changes detection vs. Competing unsupervised approaches was evaluated on Mean F1 score detection performance. An unsupervised segmentation method based on abrupt changes detection achieved mean F1 scores of 91.2%, 94.3%, and 96.3% across three databases of ~3,500 heart sound signals.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: