From the Department of Anesthesiology and Perioperative Care (C.K.L., M.C.) Department of Computer Sciences (C.K.L., P.B.) Department of Bioengineering (M.C.), University of California Irvine, Irvine, California Department of Anesthesiology and Perioperative Medicine, University of California Los Angeles, Los Angeles, California (I.H., E.G., M.C.).
Anesthesiology. 2018 Oct;129(4):649-662. doi: 10.1097/ALN.0000000000002186.
WHAT THIS ARTICLE TELLS US THAT IS NEW: BACKGROUND:: The authors tested the hypothesis that deep neural networks trained on intraoperative features can predict postoperative in-hospital mortality.
The data used to train and validate the algorithm consists of 59,985 patients with 87 features extracted at the end of surgery. Feed-forward networks with a logistic output were trained using stochastic gradient descent with momentum. The deep neural networks were trained on 80% of the data, with 20% reserved for testing. The authors assessed improvement of the deep neural network by adding American Society of Anesthesiologists (ASA) Physical Status Classification and robustness of the deep neural network to a reduced feature set. The networks were then compared to ASA Physical Status, logistic regression, and other published clinical scores including the Surgical Apgar, Preoperative Score to Predict Postoperative Mortality, Risk Quantification Index, and the Risk Stratification Index.
In-hospital mortality in the training and test sets were 0.81% and 0.73%. The deep neural network with a reduced feature set and ASA Physical Status classification had the highest area under the receiver operating characteristics curve, 0.91 (95% CI, 0.88 to 0.93). The highest logistic regression area under the curve was found with a reduced feature set and ASA Physical Status (0.90, 95% CI, 0.87 to 0.93). The Risk Stratification Index had the highest area under the receiver operating characteristics curve, at 0.97 (95% CI, 0.94 to 0.99).
Deep neural networks can predict in-hospital mortality based on automatically extractable intraoperative data, but are not (yet) superior to existing methods.
作者测试了一个假设，即基于术中特征训练的深度神经网络可以预测术后住院死亡率。
用于训练和验证算法的数据包括 59985 名患者，这些患者在手术结束时提取了 87 个特征。使用具有逻辑输出的前馈网络，通过具有动量的随机梯度下降进行训练。深度神经网络在 80%的数据上进行训练，20%的数据保留用于测试。作者通过添加美国麻醉师协会（ASA）身体状况分类和深度神经网络对简化特征集的稳健性来评估深度神经网络的改进。然后将这些网络与 ASA 身体状况、逻辑回归以及其他已发表的临床评分（包括手术阿普加评分、预测术后死亡率的术前评分、风险量化指数和风险分层指数）进行比较。
在训练集和测试集中，院内死亡率分别为 0.81%和 0.73%。具有简化特征集和 ASA 身体状况分类的深度神经网络的受试者工作特征曲线下面积最高，为 0.91（95%置信区间，0.88 至 0.93）。具有简化特征集和 ASA 身体状况的逻辑回归曲线下面积最高，为 0.90（95%置信区间，0.87 至 0.93）。风险分层指数的受试者工作特征曲线下面积最高，为 0.97（95%置信区间，0.94 至 0.99）。
深度神经网络可以根据自动提取的术中数据预测院内死亡率，但目前并不（优于）现有方法。
