Deep learning-based risk prediction for interventional clinical trials based on protocol design: A retrospective study

• DOI: 10.1016/j.patter.2023.100689
• Publication Date: 2023-02-10T16:47:32Z
• Authors: Sohrab Ferdowsi; Julien Knafou; Nikolay Borissov; David Vicente Alv...; Rahul Mishra; Poorya Amini; Douglas Teodoro
• Keywords: Ensemble Learning

Success rate of clinical trials (CTs) is low, with the protocol design itself being considered a major risk factor. We aimed to investigate use deep learning methods predict CTs based on their protocols. Considering changes and final status, retrospective assignment method was proposed label according medium, high levels. Then, transformer graph neural networks were designed combined in an ensemble model learn infer ternary categories. The achieved robust performance (area under receiving operator characteristic curve [AUROC] 0.8453 [95% confidence interval: 0.8409-0.8495]), similar individual architectures but significantly outperforming baseline bag-of-words features (0.7548 [0.7493-0.7603] AUROC). demonstrate potential predicting from protocols, paving way for customized mitigation strategies during design.