A Deep Learning Approach to Predict Bleeding Risk Over Time in Patients on Extended Anticoagulation Therapy

Background Thus far, all the clinical models developed to predict major bleeding in patients on extended anticoagulation therapy use the baseline predictors to stratify patients into different risk groups. Therefore, these models do not account for the clinical changes and events that occur after the baseline visit, which can modify risk of bleeding. However, it is difficult to develop predictive models from the routine follow-up clinical interviews which are irregular sequences of multivariate time series data. Objectives To demonstrate that deep learning can incorporate patient time-series follow-up data to improve prediction of major bleeding. Method We used the baseline and follow-up data that was collected over 8 years in a longitudinal cohort study of 2542 patients, of whom 118 had major bleeding. Four supervised neural network-based machine learning models were trained on the baseline, or the follow-up, or both datasets on 70% of the data. The performance of these models was evaluated, along with modified versions of previously developed clinical models (CHAP, ACCP, RIETE, VTE-BLEED, HAS-BLED, and OBRI), on the remaining 30% of the data. Results An ensemble of feedforward and recurrent neural networks that used the baseline and follow-up data was the best-performing model, achieving a sensitivity and a specificity of 61% and 82%, respectively, in identifying major bleeding, and it outperformed the previously developed clinical models in terms of area under the ROC curve (82%) and area under the precision-recall curve (14%). Conclusion Time series follow-up data can improve major bleeding prediction in patients on extended anticoagulation therapy.