Researchers from the University of Cambridge in the UK have developed a machine learning algorithm capable of detecting heart murmurs in dogs, a significant indicator of heart disease. This algorithm, adapted from a similar technology used for humans, analyzes audio recordings captured by digital stethoscopes, achieving an impressive accuracy rate of 90%, comparable to that of expert cardiologists. Given that heart disease is prevalent among dogs—especially small breeds like King Charles Spaniels—this innovative tool could enable veterinarians to identify heart problems earlier and administer timely treatment, ultimately enhancing the quality of life for affected pets.
Dr. Andrew McDonald, the first author of the study, emphasized the urgency of early detection in dogs, stating, “Heart disease in humans is a huge health issue, but in dogs, it’s an even bigger problem.”
He noted that most smaller breeds are likely to develop heart disease as they age, making it critical for veterinarians to recognize the signs. The research team, led by Professor Anurag Agarwal, highlighted the absence of existing databases of canine heart sounds, prompting them to use a human heart sound database for initial development. Unlike human patients, dogs cannot communicate their discomfort, making tools like this algorithm essential for timely diagnosis and treatment.
The findings were published in the Journal of Veterinary Internal Medicine, showcasing the potential of artificial intelligence to assist veterinary professionals in delivering better care for dogs.
Peoplesmind
