+1 877 705 1914• 24 / 7
HIPAA Compliant
ISO 27001 Certified

Actively Recruiting

Age: 18Years +
All Genders
ID07795450

Validating a Pre-trained Model for Time-Series Electronic Health Records to Predict ICU Patient Outcomes

Led by Peking University People's Hospital · Updated on 2026-08-31

4800

Participants Needed

1

Research Sites

52 weeks

Total Duration

AI-Summary

What This Trial Is About

Researchers are evaluating the clinical validation of a pre-trained and fine-tuned Time-Series Electronic Health Record EHR Foundation Model designed to analyze ICU patient data. This observational study aims to test the models ability to provide accurate and dynamic prognostic support, such as predicting 28-day mortality, in the real-world clinical setting of Peking University Peoples Hospital. The study combines retrospective and prospective cohort designs to assess the models generalization and predictive performance. Participants include ICU patients aged 18 years and older with stays of at least 24 hours and complete electronic health records. The study identifies two groups those who survive beyond 28 days and those who do not, with outcomes obtained from hospital records or telephone follow-ups if hospital stays are short. This approach allows researchers to evaluate the models prognostic accuracy in actual clinical practice. During the study, participants electronic health data, including demographics and laboratory tests, are collected and monitored. Outcomes such as in-hospital mortality and discharge status are recorded, with follow-up calls made when necessary to confirm 28-day survival. The research team measures how well the EHR foundation model predicts these outcomes to support clinical decision-making in critical care settings. The study is scheduled to continue through December 2027.

CONDITIONS

Brief Title

Clinical Validation of a Pre-trained and Fine-tuned Model for Analyzing Time-Series Electronic Health Records in Critical Care

Research Team

H

Huiying Zhao, PhD

X

Xiaojiang Liu, M.Med

Not the Right Trial for You?

Explore thousands of other clinical trials that might be a better match.
Sign up to get personalized trial recommendations delivered to your inbox.

Already have an account? Log in here