ContrastLOS: A Graph-Based Deep Learning Model With Contrastive Pre-Training for Improved ICU Length-of-Stay Prediction

Accurate prediction of intensive care unit (ICU) length of stay (LOS) is crucial for optimizing resource allocation and improving patient outcomes. We propose ContrastLOS, a novel graph-based deep learning model that integrates graph transformer networks with contrastive pre-training to enhance ICU...

Full description

Saved in:
Bibliographic Details
Main Authors: Guangrui Fan, Aixiang Liu, Chao Zhang
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10883945/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849722721922973696
author Guangrui Fan
Aixiang Liu
Chao Zhang
author_facet Guangrui Fan
Aixiang Liu
Chao Zhang
author_sort Guangrui Fan
collection DOAJ
description Accurate prediction of intensive care unit (ICU) length of stay (LOS) is crucial for optimizing resource allocation and improving patient outcomes. We propose ContrastLOS, a novel graph-based deep learning model that integrates graph transformer networks with contrastive pre-training to enhance ICU LOS prediction. By constructing dynamic patient similarity graphs, ContrastLOS captures intricate inter-patient relationships and temporal dependencies. Through contrastive learning on unlabeled data, the model develops robust patient embeddings suitable for both classification and regression tasks. Experiments on three comprehensive ICU datasets—MIMIC-III, MIMIC-IV v3.0, and eICU—show that ContrastLOS consistently outperforms state-of-the-art baselines. For classification (predicting whether LOS exceeds 3 or 7 days), it achieves AUROC scores of up to 82.1%, improving performance by up to 5.2% compared to competitive methods. In regression tasks, ContrastLOS attains the lowest RMSE (2.41 days) and MAE (1.48 days) on the multi-center eICU dataset. Notably, it maintains an AUROC of 76.8% with only 10% labeled data, highlighting its effectiveness in low-resource settings. These findings suggest that ContrastLOS can serve as a robust clinical decision support tool for critical care management.
format Article
id doaj-art-d8115f6690574ea0a326228799b69ff3
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-d8115f6690574ea0a326228799b69ff32025-08-20T03:11:15ZengIEEEIEEE Access2169-35362025-01-0113341323414810.1109/ACCESS.2025.354089610883945ContrastLOS: A Graph-Based Deep Learning Model With Contrastive Pre-Training for Improved ICU Length-of-Stay PredictionGuangrui Fan0https://orcid.org/0009-0001-8570-1636Aixiang Liu1https://orcid.org/0000-0001-5621-2502Chao Zhang2Department of Computer Science and Technology, Taiyuan University of Science and Technology, Wanbailin, Taiyuan, Shanxi, ChinaDepartment of Health Information Management, Fenyang College, Shanxi Medical University, Fenyang, Shanxi, ChinaDepartment of Health Information Management, Fenyang College, Shanxi Medical University, Fenyang, Shanxi, ChinaAccurate prediction of intensive care unit (ICU) length of stay (LOS) is crucial for optimizing resource allocation and improving patient outcomes. We propose ContrastLOS, a novel graph-based deep learning model that integrates graph transformer networks with contrastive pre-training to enhance ICU LOS prediction. By constructing dynamic patient similarity graphs, ContrastLOS captures intricate inter-patient relationships and temporal dependencies. Through contrastive learning on unlabeled data, the model develops robust patient embeddings suitable for both classification and regression tasks. Experiments on three comprehensive ICU datasets—MIMIC-III, MIMIC-IV v3.0, and eICU—show that ContrastLOS consistently outperforms state-of-the-art baselines. For classification (predicting whether LOS exceeds 3 or 7 days), it achieves AUROC scores of up to 82.1%, improving performance by up to 5.2% compared to competitive methods. In regression tasks, ContrastLOS attains the lowest RMSE (2.41 days) and MAE (1.48 days) on the multi-center eICU dataset. Notably, it maintains an AUROC of 76.8% with only 10% labeled data, highlighting its effectiveness in low-resource settings. These findings suggest that ContrastLOS can serve as a robust clinical decision support tool for critical care management.https://ieeexplore.ieee.org/document/10883945/Contrastive pre-traininggraph-based learninghealthcare AIintensive care unit (ICU)length of stay (LOS) prediction
spellingShingle Guangrui Fan
Aixiang Liu
Chao Zhang
ContrastLOS: A Graph-Based Deep Learning Model With Contrastive Pre-Training for Improved ICU Length-of-Stay Prediction
IEEE Access
Contrastive pre-training
graph-based learning
healthcare AI
intensive care unit (ICU)
length of stay (LOS) prediction
title ContrastLOS: A Graph-Based Deep Learning Model With Contrastive Pre-Training for Improved ICU Length-of-Stay Prediction
title_full ContrastLOS: A Graph-Based Deep Learning Model With Contrastive Pre-Training for Improved ICU Length-of-Stay Prediction
title_fullStr ContrastLOS: A Graph-Based Deep Learning Model With Contrastive Pre-Training for Improved ICU Length-of-Stay Prediction
title_full_unstemmed ContrastLOS: A Graph-Based Deep Learning Model With Contrastive Pre-Training for Improved ICU Length-of-Stay Prediction
title_short ContrastLOS: A Graph-Based Deep Learning Model With Contrastive Pre-Training for Improved ICU Length-of-Stay Prediction
title_sort contrastlos a graph based deep learning model with contrastive pre training for improved icu length of stay prediction
topic Contrastive pre-training
graph-based learning
healthcare AI
intensive care unit (ICU)
length of stay (LOS) prediction
url https://ieeexplore.ieee.org/document/10883945/
work_keys_str_mv AT guangruifan contrastlosagraphbaseddeeplearningmodelwithcontrastivepretrainingforimprovediculengthofstayprediction
AT aixiangliu contrastlosagraphbaseddeeplearningmodelwithcontrastivepretrainingforimprovediculengthofstayprediction
AT chaozhang contrastlosagraphbaseddeeplearningmodelwithcontrastivepretrainingforimprovediculengthofstayprediction