QTNet: Predicting Drug-Induced QT prolongation with Deep Neural Networks

BackgroundDrug-induced QTc prolongation (diQTP) is frequent and associated with a risk of sudden cardiac death. Identifying patients at risk of diQTP can enhance monitoring and treatment plans. ObjectiveTo develop a machine learning architecture for prediction of extreme diQTP (QTc >500ms OR {Delta}QTc >60ms) at the onset of treatment with a QTc prolonging drug. MethodsWe included 4,628 adult patients who received a baseline ECG within 6 months prior to treatment onset with a QTc prolonging drug and a follow-up ECG after the fifth dose. We collected known clinical QTc prolongation risk factors (CF). We developed a novel neural network architecture (QTNet) to predict diQTP from both the CF and baseline ECG data (ECGD), composed of both the ECG waveform and measurements (i.e. QTc),by fusing a state-of-the-art convolution neural network to process raw ECG waveforms with the CF using amulti-layer perceptron. We fit a logistic regression model using the CF, replicating RISQ-PATH as Baseline. We further compared the performance of QTNet (CF+ECGD) to neural network models trained using three variable subsets: a) baseline QTc (QTC-NN), b) CF-NN, and c) ECGD-NN. ResultsdiQTP was present in 1030 patients (22.3%), of which baseline QTc was normal (QTc View larger version (28K): org.highwire.dtl.DTLVardef@1bbd894org.highwire.dtl.DTLVardef@1880364org.highwire.dtl.DTLVardef@96a754org.highwire.dtl.DTLVardef@c394b7_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOFigure.C_FLOATNO Performance of QTNet predicting diQTP. Compared models include the Baseline CF logistic regression model (Baseline), Reference QTc (QTC), NN trained on clinical features (CF-NN) and NN trained on ECG data (ECGD-NN). diQTP, drug induced QT prolongation; CF, clinical risk factors; NN, neural network. C_FIG ConclusionWe developed QTNet, the first deep learning model for predicting extreme diQTP, outperforming models trained on known clinical risk factors.