Multi-Model Cardiac Screening Using PCG And ECG Signals

DOI

10.2991/978-94-6239-693-7_12How to use a DOI?

Keywords

Phonocardiogram (PCG); Electrocardiogram (ECG); Gradient Boosting; Random Forest; Multi-Label Classification; SHAP interpretability; biomarker analysis

Abstract

Cardiac evaluation at initial stages depends on precise handling of biological data. A method involving separate models appears here, examining PCG and ECG signals through distinct computational paths. Instead of merging information early, processing stays independent, allowing use even if only one signal type exists. From the CirCor collection, heart audio recordings feed into the PCG pathway. After signal processing, features tied to timing and sound properties undergo extraction. Following this step, classification relies on a Gradient Boosting framework. This part identifies abnormal sounds, estimate valve location, assess severity levels, and calculates murmur likelihood scores. Analysis of heart signals uses data drawn from the PTB-XL collection, applying a Random Forest method to assess irregularities and perform analysis across rhythm, waveform form, subclass, superclass, and severity levels. Insights into model decisions emerge via SHAP-driven examination, scrutiny of biological markers, graphical displays of waveforms, together with spectrogram visualizations for PCG and R-R-interval analysis for ECG signals. Access occurs through an online platform requiring only a patient ID, producing distinct reports: one for phonocardiogram findings, another for electrocardiographic results. With focus placed on clarity and separation of sensing channels, the framework supports usable, supporting automated and interpretable cardiac screening.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - M. Shridhani
AU  - V. Subapriya
PY  - 2026
DA  - 2026/06/16
TI  - Multi-Model Cardiac Screening Using PCG And ECG Signals
BT  - Proceedings of the International Conference on Intelligent Systems for a Sustainable Future (ISSF 2026)
PB  - Atlantis Press
SP  - 104
EP  - 111
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6239-693-7_12
DO  - 10.2991/978-94-6239-693-7_12
ID  - Shridhani2026
ER  -