Artificial Intelligence Deployed for Canine Heart Murmur Detection

Scientists have announced their application of Artificial Intelligence (AI) for the precise identification of heart murmurs in canines. A group of researchers at the University of Cambridge created an algorithm—defined as a series of coded instructions—initially intended for human use, which has proven capable of detecting and classifying murmurs in dogs. Heart murmurs represent a primary sign of cardiac disease, a condition prevalent among numerous smaller dog breeds, such as King Charles Spaniels. Dr. Andrew McDonald, who served as the lead author of this study, stated: “Heart disease in humans is a huge health issue, but in dogs it’s an even bigger problem.” He further commented: “As far as we’re aware, there are no existing databases of heart sounds in dogs, which is why we started out with a database of heart sounds in humans.” He concluded by noting: “Mammalian hearts are fairly similar, and when things go wrong, they tend to go wrong in similar ways.” The research team initially utilized a database comprising heart sounds from approximately 1,000 human patients. From this, they developed a machine-learning algorithm designed to mimic a cardiologist’s detection of a heart murmur. Subsequently, this algorithm underwent adaptation to enable its application with canine heart sounds. For the study, nearly 800 dogs undergoing standard heart check-ups at four specialized veterinary centers throughout the UK participated. These animals underwent a comprehensive physical examination and a heart scan performed by a cardiologist to assess any heart murmurs and diagnose cardiac conditions. Their heart rhythms were documented using an electronic stethoscope. This collection of data is considered to be the most extensive dataset of dog heart sounds ever compiled. Co-author Professor Jose Novo Matos stated that the team aimed to gather data “from dogs of all shapes, sizes and ages.” He further commented: “The more data we have to train it, the more useful our algorithm will be, both for vets and for dog owners.” An evaluation of the algorithm revealed its concordance with a cardiologist’s assessment in more than half of instances. Furthermore, in 90% of cases, its grading was within one grade of the cardiologist’s evaluation, a finding that researchers deemed encouraging. Follow Cambridgeshire news on BBC Sounds, Facebook, Instagram and X.Copyright 2024 BBC. All rights reserved. The BBC is not responsible for the content of external sites. Read about our approach to external linking.