DeepJoint: Robust Survival Modelling Under Clinical Presence Shift

Vincent Jeanselme, Glen Martin, Niels Peek, Matthew Sperrin, Brian Tom, Jessica Barrett

Keywords: Electronic Health Records, Survival, Multi-task, Joint Model

TL;DR: This paper proposes a multi-task neural network to tackle the problem of informative sampling in electronic health records.

Abstract: Medical data arise from the complex interaction between patients and healthcare systems. This data-generating process often constitutes an informative process. Prediction models often ignore this process or only partially leverage it, potentially hampering performance and transportability when this interaction evolves. This work explores how current models may suffer from shifts in this clinical presence process and proposes a multi-task recurrent neural network to tackle this issue. The proposed joint modelling competes with state-of-the-art predictive models on a real-world prediction task. More importantly, the approach appears more robust to change in the clinical presence setting. This analysis emphasises the importance of modelling clinical presence to improve performance and transportability.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2205.13481/code)