Saved in:
Bibliographic Details
Main Authors: Xu, Ran, Lu, Yiwen, Liu, Chang, Chen, Yong, Sun, Yan, Hu, Xiao, Ho, Joyce C, Yang, Carl
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.05682
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909219856318464
author Xu, Ran
Lu, Yiwen
Liu, Chang
Chen, Yong
Sun, Yan
Hu, Xiao
Ho, Joyce C
Yang, Carl
author_facet Xu, Ran
Lu, Yiwen
Liu, Chang
Chen, Yong
Sun, Yan
Hu, Xiao
Ho, Joyce C
Yang, Carl
contents Electronic Health Records (EHRs) contain rich patient information and are crucial for clinical research and practice. In recent years, deep learning models have been applied to EHRs, but they often rely on massive features, which may not be readily available for all patients. We propose HTP-Star, which leverages hypergraph structures with a pretrain-then-finetune framework for modeling EHR data, enabling seamless integration of additional features. Additionally, we design two techniques, namely (1) Smoothness-inducing Regularization and (2) Group-balanced Reweighting, to enhance the model's robustness during fine-tuning. Through experiments conducted on two real EHR datasets, we demonstrate that HTP-Star consistently outperforms various baselines while striking a balance between patients with basic and extra features.
format Preprint
id arxiv_https___arxiv_org_abs_2406_05682
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle From Basic to Extra Features: Hypergraph Transformer Pretrain-then-Finetuning for Balanced Clinical Predictions on EHR
Xu, Ran
Lu, Yiwen
Liu, Chang
Chen, Yong
Sun, Yan
Hu, Xiao
Ho, Joyce C
Yang, Carl
Machine Learning
Artificial Intelligence
Electronic Health Records (EHRs) contain rich patient information and are crucial for clinical research and practice. In recent years, deep learning models have been applied to EHRs, but they often rely on massive features, which may not be readily available for all patients. We propose HTP-Star, which leverages hypergraph structures with a pretrain-then-finetune framework for modeling EHR data, enabling seamless integration of additional features. Additionally, we design two techniques, namely (1) Smoothness-inducing Regularization and (2) Group-balanced Reweighting, to enhance the model's robustness during fine-tuning. Through experiments conducted on two real EHR datasets, we demonstrate that HTP-Star consistently outperforms various baselines while striking a balance between patients with basic and extra features.
title From Basic to Extra Features: Hypergraph Transformer Pretrain-then-Finetuning for Balanced Clinical Predictions on EHR
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2406.05682