Paper ID: 2212.12067

Enhancing the prediction of disease outcomes using electronic health records and pretrained deep learning models

Zhichao Yang, Weisong Liu, Dan Berlowitz, Hong Yu

Question: Can an encoder-decoder architecture pretrained on a large dataset of longitudinal electronic health records improves patient outcome predictions? Findings: In this prognostic study of 6.8 million patients, our denoising sequence-to-sequence prediction model of multiple outcomes outperformed state-of-the-art models scuh pretrained BERT on a broad range of patient outcomes, including intentional self-harm and pancreatic cancer. Meaning: Deep bidirectional and autoregressive representation improves patient outcome prediction.

Submitted: Dec 22, 2022