EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Understanding uncertainty in large language model predictions of early death in critically ill patients: a conformal prediction approach

Fatemeh Shah-Mohammadi iD, Alexander Millar iD, Julio C Facelli iD, Ramkiran Gouripeddi iD

JAMIA Open · 05.05.2026 · 9 (3)

DOI: 10.1093/jamiaopen/ooag108 ↗

Kurz gefasst

Fachartikel in der Zeitschrift JAMIA Open, online erschienen am 19.06.2026. Verfasst von Fatemeh Shah-Mohammadi und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biomedical Informatics, University of Utah, Salt Lake City, UT 84108, United States. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCATS NIH HHS, National Institutes of Health.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Early prediction of in-hospital death remains a significant challenge due to the limited availability of structured data during initial admission. Unstructured clinical notes, which often contain important observations and impressions, are an underutilized resource for real-time risk stratification. While leveraging recent advances in large language models (LLMs) is a promising approach to use this unstructured information, the lack of understanding of the uncertainty of LLM predictions, at the patient level, for such critical forecasts is a serious deterrence for their use in clinical settings.

Objective

This study aims to demonstrate the use of conformal prediction to evaluate the effectiveness and confidence in predicting in-hospital death probability for an individual patient using LLMs, specifically GPT-4o and unstructured clinical notes.

Methods

We applied conformal prediction to quantify the uncertainty of GPT-4o’s zero-shot predictions for in-hospital death, leveraging concatenated clinical notes documented from the first 24 h of intensive care unit (ICU) admission in MIMIC-III for patients with acute kidney failure who were admitted through the emergency department (ED).

Results

Across both classes “in-hospital death” and “in-hospital survive,” the GPT model performed better on the in-hospital death class, achieving precision of 0.52 (95% CI 0.48-0.56), recall of 0.93 (95% CI, 0.90-0.95), and F1-score 0.66 (95% CI, 0.63-0.70). The conformal prediction (CP) framework provided an overall empirical coverage of 90.4%, exceeding the target threshold of 90%. However, class-specific coverage was imbalanced, with 99.7% for the death and 81.1% for the survived class.

Conclusions

The model’s outputs exhibit overconfidence, particularly in cases of incorrect predictions. Integrating conformal prediction provides a promising approach to quantifying and calibrating uncertainty in large language model outputs for individual patient predictions, thereby enhancing their potential applicability for clinical decision-making.

Schlagwörter

Stichwörter der Autor:innen

GPTconformal predictionemergency departmentin-hospital deathlarge language model

Autor:innen und Institutionen

  • Department of Biomedical Informatics, University of Utah, Salt Lake City, UT 84108, United States.
  • Utah Clinical and Translational Science Institute, University of Utah, Salt Lake City, UT 84108, United States.

ORCID-Profile: Fatemeh Shah-Mohammadi, Alexander Millar, Julio C Facelli, Ramkiran Gouripeddi

Förderung

  • NCATS NIH HHS
  • National Institutes of Health

Zitieren

APA

Shah-Mohammadi, F., Millar, A., C Facelli, J., & Gouripeddi, R. (2026). Understanding uncertainty in large language model predictions of early death in critically ill patients: a conformal prediction approach. JAMIA Open, 9(3). https://doi.org/10.1093/jamiaopen/ooag108

Vancouver

Shah-Mohammadi F, Millar A, C Facelli J, Gouripeddi R. Understanding uncertainty in large language model predictions of early death in critically ill patients: a conformal prediction approach. JAMIA Open. 2026;9(3). doi:10.1093/jamiaopen/ooag108

BibTeX
@article{ShahMohammadi2026,
  author = {Shah-Mohammadi, Fatemeh and Millar, Alexander and C Facelli, Julio and Gouripeddi, Ramkiran},
  title = {Understanding uncertainty in large language model predictions of early death in critically ill patients: a conformal prediction approach},
  journal = {JAMIA Open},
  year = {2026},
  volume = {9},
  number = {3},
  doi = {10.1093/jamiaopen/ooag108}
}
Kostenlos starten →