Department of Computer Science, Stanford University, Palo Alto, CA 94025.
Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center, Los Angeles, CA, 90048.
EBioMedicine. 2021 Nov;73:103613. doi: 10.1016/j.ebiom.2021.103613. Epub 2021 Oct 14.
Laboratory testing is routinely used to assay blood biomarkers to provide information on physiologic state beyond what clinicians can evaluate from interpreting medical imaging. We hypothesized that deep learning interpretation of echocardiogram videos can provide additional value in understanding disease states and can evaluate common biomarkers results.
We developed EchoNet-Labs, a video-based deep learning algorithm to detect evidence of anemia, elevated B-type natriuretic peptide (BNP), troponin I, and blood urea nitrogen (BUN), as well as values of ten additional lab tests directly from echocardiograms. We included patients (n = 39,460) aged 18 years or older with one or more apical-4-chamber echocardiogram videos (n = 70,066) from Stanford Healthcare for training and internal testing of EchoNet-Lab's performance in estimating the most proximal biomarker result. Without fine-tuning, the performance of EchoNet-Labs was further evaluated on an additional external test dataset (n = 1,301) from Cedars-Sinai Medical Center. We calculated the area under the curve (AUC) of the receiver operating characteristic curve for the internal and external test datasets.
On the held-out test set of Stanford patients not previously seen during model training, EchoNet-Labs achieved an AUC of 0.80 (0.79-0.81) in detecting anemia (low hemoglobin), 0.86 (0.85-0.88) in detecting elevated BNP, 0.75 (0.73-0.78) in detecting elevated troponin I, and 0.74 (0.72-0.76) in detecting elevated BUN. On the external test dataset from Cedars-Sinai, EchoNet-Labs achieved an AUC of 0.80 (0.77-0.82) in detecting anemia, of 0.82 (0.79-0.84) in detecting elevated BNP, of 0.75 (0.72-0.78) in detecting elevated troponin I, and of 0.69 (0.66-0.71) in detecting elevated BUN. We further demonstrate the utility of the model in detecting abnormalities in 10 additional lab tests. We investigate the features necessary for EchoNet-Labs to make successful detection and identify potential mechanisms for each biomarker using well-known and novel explainability techniques.
These results show that deep learning applied to diagnostic imaging can provide additional clinical value and identify phenotypic information beyond current imaging interpretation methods.
J.W.H. and B.H. are supported by the NSF Graduate Research Fellowship. D.O. is supported by NIH K99 HL157421-01. J.Y.Z. is supported by NSF CAREER 1942926, NIH R21 MD012867-01, NIH P30AG059307 and by a Chan-Zuckerberg Biohub Fellowship.
实验室检测常用于检测血液生物标志物，以提供超出临床医生通过解读医学影像评估范围的生理状态信息。我们假设超声心动图视频的深度学习解释可以提供额外的价值，以帮助理解疾病状态，并评估常见的生物标志物结果。
我们开发了 EchoNet-Labs，这是一种基于视频的深度学习算法，可直接从超声心动图中检测贫血、升高的 B 型利钠肽（BNP）、肌钙蛋白 I 和血尿素氮（BUN）的证据，以及另外 10 项实验室检测的结果。我们纳入了来自斯坦福健康医疗中心的年龄在 18 岁或以上的一名或多名心尖 4 腔超声心动图视频（n=70066）的患者（n=39460）进行训练，并用于内部测试 EchoNet-Lab 对估计最接近的生物标志物结果的性能。未经微调，EchoNet-Labs 的性能在雪松西奈医疗中心的另一个外部测试数据集（n=1301）上进行了进一步评估。我们计算了内部和外部测试数据集的受试者工作特征曲线下面积（AUC）。
在斯坦福患者的预留测试集中，该测试集未在模型训练期间看到，EchoNet-Labs 在检测贫血（低血红蛋白）方面的 AUC 为 0.80（0.79-0.81），在检测升高的 BNP 方面的 AUC 为 0.86（0.85-0.88），在检测升高的肌钙蛋白 I 方面的 AUC 为 0.75（0.73-0.78），在检测升高的 BUN 方面的 AUC 为 0.74（0.72-0.76）。在雪松西奈的外部测试数据集中，EchoNet-Labs 在检测贫血方面的 AUC 为 0.80（0.77-0.82），在检测升高的 BNP 方面的 AUC 为 0.82（0.79-0.84），在检测升高的肌钙蛋白 I 方面的 AUC 为 0.75（0.72-0.78），在检测升高的 BUN 方面的 AUC 为 0.69（0.66-0.71）。我们进一步展示了该模型在检测 10 项额外实验室检测异常方面的效用。我们研究了 EchoNet-Labs 成功检测所需的特征，并使用众所周知和新颖的可解释性技术确定了每个生物标志物的潜在机制。
这些结果表明，应用于诊断成像的深度学习可以提供额外的临床价值，并识别超出当前成像解释方法的表型信息。
J.W.H.和 B.H. 得到了 NSF 研究生研究奖学金的支持。D.O. 得到了 NIH K99 HL157421-01 的支持。J.Y.Z. 得到了 NSF CAREER 1942926、NIH R21 MD012867-01、NIH P30AG059307 和 Chan-Zuckerberg Biohub 奖学金的支持。
Zotero 插件
