Researchers have developed a groundbreaking machine learning algorithm that accurately detects heart murmurs in dogs, a key sign of cardiac disease affecting many smaller breeds like King Charles Spaniels.
Led by the University of Cambridge, the team adapted an algorithm originally designed for humans to identify and grade heart murmurs from audio recordings taken with digital stethoscopes. Impressively, the algorithm achieved a 90% sensitivity rate, matching expert cardiologists.
Heart murmurs signal mitral valve disease, the most common heart condition in adult dogs, affecting about one in 30 dogs, especially smaller and older ones.
Early detection is crucial, as timely treatment can significantly extend their lives. This innovative technology could serve as an affordable screening tool for veterinarians, enhancing the quality of life for our canine companions.
“Heart disease in humans is a huge health issue, but in dogs, it’s an even bigger problem,” said first author Dr Andrew McDonald from Cambridge’s Department of Engineering. “Most smaller dog breeds will have heart disease when they get older, but obviously dogs can’t communicate in the same way that humans can, so it’s up to primary care vets to detect heart disease early enough so it can be treated.”
“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,” said Professor Anurag Agarwal, who led the research. “Mammalian hearts are fairly similar, and when things go wrong, they tend to go wrong in similar ways.”
Researchers initiated their groundbreaking work with an extensive database of heart sounds from approximately 1,000 human patients, developing a sophisticated machine learning algorithm capable of replicating a cardiologist’s detection of heart murmurs. They then ingeniously adapted this algorithm for use with canine heart sounds.
To achieve this, the researchers meticulously gathered data from nearly 800 dogs undergoing routine heart examinations across four veterinary specialist centers in the UK. Each dog benefited from a thorough physical evaluation and an echocardiogram by a trained cardiologist, allowing for the grading of any heart murmurs and identification of cardiac diseases. This effort has resulted in the largest dataset of dog heart sounds ever compiled, setting a new standard in veterinary cardiology.
“Mitral valve disease mainly affects smaller dogs, but to test and improve our algorithm, we wanted to get data from dogs of all shapes, sizes, and ages,” said co-author Professor Jose Novo Matos from Cambridge’s Department of Veterinary Medicine, a specialist in small animal cardiology. “The more data we have to train it, the more useful our algorithm will be, both for vets and for dog owners.”
The researchers have successfully refined an algorithm designed to detect and grade heart murmurs through audio recordings. This innovative tool not only distinguishes between murmurs linked to mild conditions and those indicative of serious heart disease requiring intervention but also holds the promise of enhancing veterinary care.
“Grading a heart murmur and determining whether the heart disease needs treatment requires a lot of experience, referral to a veterinary cardiologist, and expensive specialized heart scans,” said Novo Matos. “We want to empower general practitioners to detect heart disease and assess its severity to help owners make the best decisions for their dogs.”
The algorithm demonstrated impressive performance, aligning with the cardiologist’s assessments in over half of the cases, and in 90% of instances, it was only one grade off from the cardiologist’s evaluation. This finding is particularly encouraging, considering the frequent discrepancies among veterinarians in grading heart murmurs.
“The grade of heart murmur is a useful differentiator for determining next steps and treatments, and we’ve automated that process,” said McDonald. “For vets and nurses without as much stethoscope skill and even those who are incredibly skilled with a stethoscope, we believe this algorithm could be a highly valuable tool.”
In humans with valve disease, the only treatment is surgery, but for dogs, effective medication is available. “Knowing when to medicate is so important, in order to give dogs the best quality of life possible for as long as possible,” said Agarwal. “We want to empower vets to help make those decisions.”
“So many people talk about AI as a threat to jobs, but for me, I see it as a tool that will make me a better cardiologist,” said Novo Matos. “We can’t perform heart scans on every dog in this country – we just don’t have enough time or specialists to screen every dog with a murmur. But tools like these could help vets and owners, so we can quickly identify those dogs who are most in need of treatment.”
Journal reference:
- Andrew McDonald, Jose Novo Matos, Joel Silva, Catheryn Partington, Eve J. Y. Lo, Virginia Luis Fuentes, Lara Barron, Penny Watson, Anurag Agarwal. A machine-learning algorithm to grade heart murmurs and stage preclinical myxomatous mitral valve disease in dogs. Journal of Veterinary Internal Medicine, 2024; DOI: 10.1111/jvim.17224