AI is detecting heart disease in dogs with the accuracy of cardiologists

Photo: Toowongsao/Adobe Stock

A team of researchers from the University of Cambridge has designed a machine learning algorithm to effectively detect and grade heart murmurs in canine patients, using audio recordings from electronic stethoscopes. Originally designed for humans, the algorithm identified heart murmurs of any grade with a sensitivity of 87.9% in testing, according to the study’s authors.¹

"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," Anurag Agarwal, PhD, head of the study, said in a news release.² "Mammalian hearts are fairly similar, and when things go wrong, they tend to go wrong in similar ways."

The team began with a database of heart sounds from approximately 1000 human patients and developed a machine learning algorithm to replicate the detection of heart murmurs by cardiologists. They then adjusted the algorithm to analyze heart sounds from dogs.

Data were collected from nearly 800 dogs undergoing routine heart checks at 4 veterinary specialist centers in the United Kingdom.² Each dog in the study received a full physical exam and an echocardiogram from a cardiologist, who graded any murmurs and diagnosed cardiac disease, while heart sounds were recorded with an electronic stethoscope.

"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," Jose Novo Matos, DVM, PhD, MSc, DECVIM (Cardiology), AFHEA, coauthor of the study and a professor in Cambridge’s department of veterinary medicine, said in the release.²

"The more data we have to train it, the more useful our algorithm will be, both for vets and for dog owners," he continued.² According to the researchers, the study, published in the Journal of Veterinary Internal Medicine, is the largest database of compiled annotated dog heart sound recordings.^1,2

Not only can the artificial intelligence (AI) program detect and rate heart murmurs from audio recordings, it can also tell the difference between murmurs linked to mild disease and those indicating more serious heart problems that need further treatment, according to the release.²

"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," Novo Matos explained in the release.² "We want to empower general practitioners to detect heart disease and assess its severity to help owners make the best decisions for their dogs."

An analysis of the algorithm's performance found that the artificial intelligence (AI) concurred with the cardiologist's evaluations in more than half of cases. In 90% of cases, it was within one grade of the cardiologist's rating.

Heart murmurs, mitral valve disease, and an AI solution

A heart murmur is a primary indicator of mitral valve disease, which is the leading heart condition in adult dogs.² Approximately 1 in 30 dogs seen by a veterinarian is found to have a heart murmur, particularly in small and older breeds where prevalence is higher.

Early detection is crucial for extending canines’ lifespan through timely treatment, given the prevalence of mitral valve disease and other heart conditions in these animals. "Heart disease in humans is a huge health issue, but in dogs it's an even bigger problem," Andrew McDonald, PhD, MEng, first author of the study, explained.² "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."

"The grade of heart murmur is a useful differentiator for determining next steps and treatments, and we've automated that process," McDonald said.² "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.”

Currently, the only treatment for humans with valve disease is surgery. For dogs however, medication is a potential solution. “Knowing when to medicate is so important in order to give dogs the best quality of life possible for as long as possible," Agarwal noted in the release.² "We want to empower vets to help make those decisions."

The new AI technology may provide a cost-effective and efficient screening solution for primary care veterinarians, improving the quality of life for dogs. "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," Novo Matos said in the release.² "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."

References

1. McDonald A, Novo Matos J, Silva J, et al. A machine‐learning algorithm to Grade Heart Murmurs and stage preclinical myxomatous mitral valve disease in dogs. Journal of Veterinary Internal Medicine. Published online October 21, 2024. https://doi.org/10.1111/jvim.17224
2. AI algorithm accurately detects heart disease in dogs. News release. ScienceDaily. October 28, 2024. Accessed October 30, 2024. https://www.sciencedaily.com/releases/2024/10/241028211501.htm