Patient Outcome and Zero-shot Diagnosis Prediction with Hypernetwork-guided Multitask Learning

TL;DR

This paper introduces a hypernetwork-guided multitask learning approach for patient outcome prediction from clinical notes, enhancing generalizability and zero-shot diagnosis prediction, especially for rare or unseen diseases.

Contribution

The paper proposes a novel hypernetwork-based method that generates task-specific parameters and incorporates semantic information to improve multitask learning and zero-shot diagnosis prediction.

Findings

Outperforms strong baselines on real-world MIMIC data

Enhances zero-shot prediction for unseen diagnoses

Improves multitask patient outcome prediction accuracy

Abstract

Multitask deep learning has been applied to patient outcome prediction from text, taking clinical notes as input and training deep neural networks with a joint loss function of multiple tasks. However, the joint training scheme of multitask learning suffers from inter-task interference, and diagnosis prediction among the multiple tasks has the generalizability issue due to rare diseases or unseen diagnoses. To solve these challenges, we propose a hypernetwork-based approach that generates task-conditioned parameters and coefficients of multitask prediction heads to learn task-specific prediction and balance the multitask learning. We also incorporate semantic task information to improves the generalizability of our task-conditioned multitask model. Experiments on early and discharge notes extracted from the real-world MIMIC database show our method can achieve better performance on multitask patient outcome prediction than strong baselines in most cases. Besides, our method can effectively handle the scenario with limited information and improve zero-shot prediction on unseen diagnosis categories.