Automated detection of pediatric congenital heart disease from phonocardiograms using deep and handcrafted feature fusion

• Published in: Computers in biology and medicine Vol. 197; no. Pt A; p. 110993
• Main Authors: Jabbar, Abdul, Grooby, Ethan, Poh, Yang Yi, Ahmad, Khawza I., Hassanuzzaman, Md, Mostafa, Raqibul, Khandoker, Ahsan H., Marzbanrad, Faezeh
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.10.2025
• Subjects: Adolescent; Child; Child, Preschool; Congenital heart disease (CHD); Deep feature fusion; Deep Learning; Diagnosis, Computer-Assisted - methods; Female; Heart Defects, Congenital - diagnosis; Heart Defects, Congenital - diagnostic imaging; Heart Defects, Congenital - physiopathology; Heart sounds; Humans; Infant; Infant, Newborn; Internal Medicine; Male; Mel-frequency cepstral coefficients (MFCC); Other; Phonocardiography (PCG); Phonocardiography - methods; Signal Processing, Computer-Assisted; Deep learning; Heart sounds; Phonocardiography (PCG); Deep feature fusion; Mel-frequency cepstral coefficients (MFCC); Congenital heart disease (CHD)
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2025.110993

Congenital heart disease (CHD) is the most common type of birth defect, impacting about 1% of live births worldwide. Echocardiography, the gold-standard diagnostic method, is costly and inaccessible in low-resource settings. Diagnosis is delayed due to limited skilled experts, whose ability to interpret pathological patterns varies significantly, causing inter- and intra-clinician variability. Therefore, we present a new method for a more accessible diagnostic modality, the digital stethoscope, to detect CHDs. Our method is based on deep feature fusion, integrating deep and handcrafted features for the automated early detection of CHDs. For this work, Phonocardiography (PCG) recordings were obtained from 751 pediatric subjects (Age:1 month- 16 years) in Bangladesh, ranging from infants to adults at four auscultation locations: mitral valve (MV), aortic valve (AV), pulmonary valve (PV), and tricuspid valve (TV). These recordings were labeled based on confirmed diagnoses by cardiologists as either cases of CHD or non-CHD. The results demonstrated that our proposed model achieved an accuracy of 92%, a sensitivity of 91%, and a specificity of 91%, based on a patient-wise split of 70% training, 20% validation, and 10% testing. Furthermore, the Area Under the Receiver Operating Characteristic curve (AUROC) of 96%, and an F1-score of 92%. This model promises efficient real-time remote detection of CHDs as a cost-effective screening tool for low-resource settings. [Display omitted] •Hybrid deep model developed for pediatric CHD detection, not just murmurs.•Enables CHD screening via digital stethoscope, affordable in low-resource areas.•Combines handcrafted and deep features for stronger feature representation.•Achieves 92% accuracy, 91% sensitivity, and 92% specificity in CHD detection. .