Cardiio, Cambridge, Massachusetts, USA.
Division of Cardiology, Department of Medicine, University of Hong Kong, Hong Kong.
Heart. 2018 Dec;104(23):1921-1928. doi: 10.1136/heartjnl-2018-313147. Epub 2018 May 31.
To evaluate the diagnostic performance of a deep learning system for automated detection of atrial fibrillation (AF) in photoplethysmographic (PPG) pulse waveforms.
We trained a deep convolutional neural network (DCNN) to detect AF in 17 s PPG waveforms using a training data set of 149 048 PPG waveforms constructed from several publicly available PPG databases. The DCNN was validated using an independent test data set of 3039 smartphone-acquired PPG waveforms from adults at high risk of AF at a general outpatient clinic against ECG tracings reviewed by two cardiologists. Six established AF detectors based on handcrafted features were evaluated on the same test data set for performance comparison.
In the validation data set (3039 PPG waveforms) consisting of three sequential PPG waveforms from 1013 participants (mean (SD) age, 68.4 (12.2) years; 46.8% men), the prevalence of AF was 2.8%. The area under the receiver operating characteristic curve (AUC) of the DCNN for AF detection was 0.997 (95% CI 0.996 to 0.999) and was significantly higher than all the other AF detectors (AUC range: 0.924-0.985). The sensitivity of the DCNN was 95.2% (95% CI 88.3% to 98.7%), specificity was 99.0% (95% CI 98.6% to 99.3%), positive predictive value (PPV) was 72.7% (95% CI 65.1% to 79.3%) and negative predictive value (NPV) was 99.9% (95% CI 99.7% to 100%) using a single 17 s PPG waveform. Using the three sequential PPG waveforms in combination (<1 min in total), the sensitivity was 100.0% (95% CI 87.7% to 100%), specificity was 99.6% (95% CI 99.0% to 99.9%), PPV was 87.5% (95% CI 72.5% to 94.9%) and NPV was 100% (95% CI 99.4% to 100%).
In this evaluation of PPG waveforms from adults screened for AF in a real-world primary care setting, the DCNN had high sensitivity, specificity, PPV and NPV for detecting AF, outperforming other state-of-the-art methods based on handcrafted features.
评估深度学习系统在光电容积脉搏波（PPG）中自动检测心房颤动（AF）的诊断性能。
我们使用来自多个公开 PPG 数据库的 149048 个 PPG 波构建的训练数据集，对一个深度卷积神经网络（DCNN）进行训练，以检测 17s PPG 波中的 AF。该 DCNN 使用来自普通门诊高 AF 风险的成年人的 3039 个智能手机获取的 PPG 波的独立测试数据集进行验证，该数据集与两名心脏病专家审查的 ECG 轨迹相对应。在相同的测试数据集上评估了基于手工特征的六个已建立的 AF 探测器，以进行性能比较。
在由 1013 名参与者（平均（SD）年龄 68.4（12.2）岁；46.8%男性）的三个连续 PPG 波组成的验证数据集中（3039 个 PPG 波），AF 的患病率为 2.8%。用于 AF 检测的 DCNN 的接收者操作特征曲线（ROC）下面积（AUC）为 0.997（95%置信区间 0.996 至 0.999），明显高于所有其他 AF 探测器（AUC 范围：0.924-0.985）。DCNN 的灵敏度为 95.2%（95%置信区间 88.3%至 98.7%），特异性为 99.0%（95%置信区间 98.6%至 99.3%），阳性预测值（PPV）为 72.7%（95%置信区间 65.1%至 79.3%），阴性预测值（NPV）为 99.9%（95%置信区间 99.7%至 100%），使用单个 17s PPG 波。使用三个连续的 PPG 波（总用时不到 1 分钟），灵敏度为 100.0%（95%置信区间 87.7%至 100%），特异性为 99.6%（95%置信区间 99.0%至 99.9%），PPV 为 87.5%（95%置信区间 72.5%至 94.9%），NPV 为 100%（95%置信区间 99.4%至 100%）。
在这项对真实世界初级保健环境中筛查 AF 的成年人的 PPG 波进行的评估中，DCNN 在检测 AF 方面具有较高的灵敏度、特异性、PPV 和 NPV，优于基于手工特征的其他最先进的方法。
