Congenital Heart Disease Detection Using Clinical Data and Auscultation Heart Sounds: a Machine Learning Approach Solange Belinha 1 , MD, Bruno Miguel Oliveira 1,2 , MSc and Pedro Pereira Rodrigues 1,2 , PhD 1 Faculty of Medicine, Department of Biochemistry, University of Porto, Al Prof Hernani Monteiro, 4200-319 Porto, Portugal 2 CINTESIS - Center for Research in Health Technologies and Information Systems, Porto, Portugal Abstract Congenital heart disease (CHD) is the most common congenital malformation and has high morbidity and mortality related to late diagnosis. Screening protocols are lacking and only 1% of murmurs are associated with CHD. The decline in auscultation skills highlights the need for better screening. This study aims to create and evaluate models for the detection of CHD using clinical data and sound features. These features were extracted using pure conventional MFCC and selected MFCC through matrix profling and motif search. Four combinations of data were used to train decision trees (DT) and artifcial neural networks (ANN), and the area under the curve (AUC) was compared. Posteriorly, models were also trained for the detection of any cardiac pathology. In both pathologies, the ANN model using clinical data and conventional MFCC showed the highest performance with AUC of 0.761 for CHD and 0.791 for any cardiac pathology. However, this is only a slight improvement when compared with the ANN models using only clinical data (0.747 and 0.789, respectively. Additionally, the inclusion of motif selected MFCC seems to worsen the model performance. Although further research is still needed, this is a potential improvement in CHD screening, particularly for primary care physicians. Keywords Heart Auscultation, Machine Learning, Congenital Heart Disease, Mel-frequency Cepstral Coefcients, Matrix Profle, Decision Tree, Artifcial Neural Network, Computer Assisted Decision 1. Introduction 1.1. Background Congenital heart disease (CHD) is the most common congenital defect in the world [1, 2] and is defned as an abnormal development of the structures of the heart and/or great vessels which is present at birth. In terms of global birth incidence, recent studies estimated a birth incidence of more than 17/1000 in 2017 [3, 4], which represents an increase of 4.2% from 1990 [5, 6]. As for global preva- lence, it is estimated that nearly 12 million people were living with CHD in 2017, representing AIxIA 2021 SMARTERCARE Workshop, November 29, 2021, Milan, IT E up201503744@up.pt (S. Belinha); boliveira@med.up.pt (B. M. Oliveira); pprodrigues@med.up.pt (P. P. Rodrigues) O 0000-0002-0760-1808 (S. Belinha); 0000−0001−7665−6506 (B. M. Oliveira); 0000-0001-7867-6682 (P. P. Rodrigues) © 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR Workshop Proceedings http://ceur-ws.org ISSN 1613-0073 CEUR Workshop Proceedings (CEUR-WS.org) 91