Ainiwaer Aikeliyaer, Hou Wen Qing, Qi Quan, Kadier Kaisaierjiang, Qin Lian, Rehemuding Rena, Mei Ming, Wang Duolao, Ma Xiang, Dai Jian Guo, Ma Yi Tong
Department of Cardiology, First Afliated Hospital of Xinjiang Medical University, State Key Laboratory of Pathogenesis, Prevention and Treatment of High Incidence Diseases in Central Asia, Urumqi, Xinjiang, 830000, China.
School of Information Network Security, Xinjiang University of Political Science and Law, Tumushuke, Xinjiang, 843802, China.
Heliyon. 2023 Dec 8;10(1):e23354. doi: 10.1016/j.heliyon.2023.e23354. eCollection 2024 Jan 15.
Due to the limitations of current methods for detecting obstructive coronary artery disease (CAD), many individuals are mistakenly or unnecessarily referred for coronary angiography (CAG).
Our goal is to create a comprehensive database of heart sounds in CAD and develop accurate deep learning algorithms to efficiently detect obstructive CAD based on heart sound signals. This will enable effective screening before undergoing CAG.
We included 320 subjects suspected of CAD who underwent CAG. We employed advanced filtering techniques and state-of-the-art deep learning models (VGG-16, 1D CNN, and ResNet18) to analyze the heart sound signals and identify obstructive CAD (defined as at least one ≥50 % stenosis). To assess the performance of our models, we prospectively recruited an additional 80 subjects for testing.
In the test set, VGG-16 exhibited the highest performance with an area under the ROC curve (AUC) of 0.834 (95 % CI, 0.736-0.930), while ResNet-18 and CNN-7 achieved AUCs of only 0.755 (95 % CI, 0.614-0.819) and 0.652 (95 % CI, 0.554-0.770) respectively. VGG-16 demonstrated a sensitivity of 80.4 % and specificity of 86.2 % in the test set. The combined diagnostic model of VGG and DF scores achieved an AUC of 0.915 (95 % CI: 0.855-0.974), and the AUC for VGG combined with PTP scores was 0.908 (95 % CI: 0.845-0.971). The sensitivity and specificity of VGG-16 exceeded 0.85 in patients with coronary artery occlusion and those with 3 vascular lesions.
Our deep learning model, based on heart sounds, offers a non-invasive and efficient screening method for obstructive CAD. It is expected to significantly reduce the number of unnecessary referrals for downstream screening.
由于目前检测阻塞性冠状动脉疾病（CAD）的方法存在局限性，许多人被错误或不必要地转诊进行冠状动脉造影（CAG）。
我们的目标是创建一个关于CAD心音的综合数据库，并开发准确的深度学习算法，以基于心音信号有效检测阻塞性CAD。这将能够在进行CAG之前进行有效的筛查。
我们纳入了320名疑似CAD并接受CAG的受试者。我们采用先进的滤波技术和最先进的深度学习模型（VGG-16、1D CNN和ResNet18）来分析心音信号并识别阻塞性CAD（定义为至少一处≥50%狭窄）。为了评估我们模型的性能，我们前瞻性地招募了另外80名受试者进行测试。
在测试集中，VGG-16表现出最高的性能，ROC曲线下面积（AUC）为0.834（95%CI，0.736-0.930），而ResNet-18和CNN-7的AUC分别仅为0.755（95%CI，0.614-0.819）和0.652（95%CI，0.554-0.770）。VGG-16在测试集中显示出80.4%的敏感性和86.2%的特异性。VGG和DF评分的联合诊断模型的AUC为0.915（95%CI：0.855-0.974），VGG与PTP评分联合的AUC为0.908（95%CI：0.845-0.971）。在冠状动脉闭塞患者和有3处血管病变的患者中，VGG-16的敏感性和特异性超过0.85。
我们基于心音的深度学习模型为阻塞性CAD提供了一种非侵入性且高效的筛查方法。预计它将显著减少下游筛查中不必要的转诊数量。
