Department of Engineering, University of Cambridge, Cambridge, United Kingdom.
Department of Veterinary Medicine, University of Cambridge, Cambridge, United Kingdom.
J Vet Intern Med. 2024 Nov-Dec;38(6):2994-3004. doi: 10.1111/jvim.17224. Epub 2024 Oct 21.
The presence and intensity of heart murmurs are sensitive indicators of several cardiac diseases in dogs, particularly myxomatous mitral valve disease (MMVD), but accurate interpretation requires substantial clinical expertise.
Assess if a machine-learning algorithm can be trained to accurately detect and grade heart murmurs in dogs and detect cardiac disease in electronic stethoscope recordings.
Dogs (n = 756) with and without cardiac disease attending referral centers in the United Kingdom.
All dogs received full physical and echocardiographic examinations by a cardiologist to grade any murmurs and identify cardiac disease. A recurrent neural network algorithm, originally trained for heart murmur detection in humans, was fine-tuned on a subset of the dog data to predict the cardiologist's murmur grade from the audio recordings.
The algorithm detected murmurs of any grade with a sensitivity of 87.9% (95% confidence interval [CI], 83.8%-92.1%) and a specificity of 81.7% (95% CI, 72.8%-89.0%). The predicted grade exactly matched the cardiologist's grade in 57.0% of recordings (95% CI, 52.8%-61.0%). The algorithm's prediction of loud or thrilling murmurs effectively differentiated between stage B1 and B2 preclinical MMVD (area under the curve [AUC], 0.861; 95% CI, 0.791-0.922), with a sensitivity of 81.4% (95% CI, 68.3%-93.3%) and a specificity of 73.9% (95% CI, 61.5%-84.9%).
A machine-learning algorithm trained on humans can be successfully adapted to grade heart murmurs in dogs caused by common cardiac diseases, and assist in differentiating preclinical MMVD. The model is a promising tool to enable accurate, low-cost screening in primary care.
心杂音的存在和强度是犬多种心脏病的敏感指标，特别是黏液样二尖瓣病变（MMVD），但准确解读需要大量临床专业知识。
评估机器学习算法是否可以训练用于准确检测和分级犬心杂音，并检测电子听诊器记录中的心脏疾病。
在英国就诊的转诊中心的有和没有心脏疾病的犬。
所有犬均由心脏病专家进行全面的体格检查和超声心动图检查，以分级任何心杂音并识别心脏疾病。最初为人类心杂音检测而训练的递归神经网络算法，在犬数据的一个子集上进行了微调，以根据音频记录预测心脏病专家的心杂音等级。
该算法检测任何等级的心杂音的敏感性为 87.9%（95%置信区间[CI]，83.8%-92.1%），特异性为 81.7%（95% CI，72.8%-89.0%）。在 57.0%的记录中（95% CI，52.8%-61.0%），预测等级与心脏病专家的等级完全匹配。该算法对响亮或刺耳的心杂音的预测能够有效区分 B1 期和 B2 期临床前 MMVD（曲线下面积[AUC]，0.861；95%CI，0.791-0.922），敏感性为 81.4%（95% CI，68.3%-93.3%），特异性为 73.9%（95% CI，61.5%-84.9%）。
在人类身上训练的机器学习算法可以成功地适应于分级犬常见心脏疾病引起的心杂音，并有助于区分临床前 MMVD。该模型是一种有前途的工具，可以在初级保健中实现准确、低成本的筛查。
