An artificial intelligence-enabled digital stethoscope demonstrates moderate murmur detection in dogs but not cats and unreliable arrhythmia classification in both species.

AI stethoscopes show moderate canine murmur detection but are unreliable in cats and for arrhythmia classification.

This prospective study evaluated the diagnostic performance of an AI-enabled digital stethoscope (EKO Core 500) for detecting cardiac murmurs and arrhythmias in dogs and cats at a university teaching hospital. Each animal underwent auscultation at four thoracic sites using the AI stethoscope, alongside evaluation by a cardiology resident, board-certified cardiologist, and fourth-year veterinary student, with 6-lead ECG and echocardiography as reference standards. In dogs, the AI stethoscope demonstrated moderate murmur detection with 86.8% sensitivity and 56.3% specificity, performing comparably to fourth-year veterinary students (κ = 0.447). However, performance in cats was markedly poor, detecting only 2 of 22 murmurs (9.1% sensitivity, κ = 0.081), likely reflecting the typically softer, more dynamic nature of feline cardiac murmurs. Murmur grade was the sole significant predictor of AI detection success, with high-grade murmurs (≥ grade 3) having 15-fold greater odds of detection. Arrhythmia classification was unreliable in both species; while the device achieved 100% sensitivity for atrial fibrillation, it failed to classify any dog as arrhythmia-free, indicating a problematic false-positive rate. These findings suggest the AI stethoscope may serve as a useful screening adjunct for canine murmurs in general practice, but should not replace trained auscultation, particularly in feline patients. Rigorous clinical validation is warranted before routine adoption of this technology in veterinary settings.

This summary was distilled by AI and may occasionally misinterpret data. Confirm critical details with the primary literature before clinical application.