EUVIMEDUpptäck EUVIMED
Multizentrische StudieValidierungsstudieOpen access

Development and Validation of a Dynamic Real-Time Risk Prediction Model for Intensive Care Units Patients Based on Longitudinal Irregular Data: Multicenter Retrospective Study

Entwicklung und Validierung eines dynamischen Echtzeit-Risiko-Vorhersagemodells für Intensivpatient:innen auf Basis longitudinaler unregelmäßiger Daten: multizentrische retrospektive Studie

Zhuo Zheng iD, Jiawei Luo iD, Yingchao Zhu iD, Lei Du iD, Lan Lan iD, Xiaobo Zhou iD, Xiaoyan Yang iD, Shixin Huang iD

Journal of Medical Internet Research · 2025-04-23 · 27 · e69293

DOI: 10.2196/69293 ↗

Auf einen Blick

Ein TBAL-Modell sagte die kurzfristige ICU-Mortalität auf Basis stündlich aktualisierter EMR-Daten vorher. Die Leistung war in MIMIC-IV und eICU-CRD hoch und blieb in der externen Validierung stabil.

Fragestellung

Kann ein interpretierbares Echtzeitmodell die ICU-Mortalität mit unregelmäßigen longitudinalen EMR-Daten verlässlich vorhersagen?

Methode

Retrospektive multizentrische Modellentwicklungs- und Validierungsstudie mit TBAL-Modell auf MIMIC-IV und eICU-CRD; externe Kreuzvalidierung und Subgruppenanalysen.

Ergebnisse

176.344 ICU-Aufenthalte. Static 12 h bis 1 Tag: AUROC 95,9 (95 %-KI 94,2–97,5) und 93,3 (95 %-KI 91,5–95,3); AUPRC 48,5 und 21,6. Dynamisch: AUROC 93,6 (95 %-KI 93,2–93,9) und 91,9 (95 %-KI 91,6–92,1).

Einschränkungen

Prospektive klinische Prüfung steht aus.

Schlussfolgerung der Autor:innen

Das TBAL-Modell bietet eine robuste, interpretierbare und generalisierbare Lösung für die dynamische Echtzeit-Vorhersage der ICU-Mortalität.

Kernaussagen

  • Stündlich aktualisierte Vorhersagen auf Basis von EMR-Daten
  • Hohe AUROCs in beiden Datenbanken und in der externen Validierung
  • Top-Prädiktoren: Laktat, Vasopressor-Einsatz, Glasgow Coma Scale

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

At a glance

Multizentrische Studie in Journal of Medical Internet Research, published online on 2025-04-23. Written by Zhuo Zheng and 7 co-authors. Institution: Department of Anesthesiology, West China Hospital of Sichuan University, Chengdu, China. Major MEDLINE topics: Intensive Care Units, Hospital Mortality. Cited 25 times so far (Crossref); 53 references. The full text is freely available in PubMed Central.

Auf einen BlickAbstractSubjectsBibliografiska uppgifterCiteRelated studies

Abstract (original language)

Background

Timely and accurate prediction of short-term mortality is critical in intensive care units (ICUs), where patients’ conditions change rapidly. Traditional scoring systems, such as the Simplified Acute Physiology Score and Acute Physiology and Chronic Health Evaluation, rely on static variables collected within the first 24 hours of admission and do not account for continuously evolving clinical states. These systems lack real-time adaptability, interpretability, and generalizability. With the increasing availability of high-frequency electronic medical record (EMR) data, machine learning (ML) approaches have emerged as powerful tools to model complex temporal patterns and support dynamic clinical decision-making. However, existing models are often limited by their inability to handle irregular sampling and missing values, and many lack rigorous external validation across institutions.

Objective

We aimed to develop a real-time, interpretable risk prediction model that continuously assesses ICU patient mortality using irregular, longitudinal EMR data, with improved performance and generalizability over traditional static scoring systems.

Methods

A time-aware bidirectional attention-based long short-term memory (TBAL) model was developed using EMR data from the MIMIC-IV (Medical Information Mart for Intensive Care) and eICU Collaborative Research Database (eICU-CRD) databases, comprising 176,344 ICU stays. The model incorporated dynamic variables, including vital signs, laboratory results, and medication data, updated hourly, to perform static and continuous mortality risk assessments. External cross-validation and subgroup sensitivity analyses were conducted to evaluate robustness and fairness. Model performance was assessed using the area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), accuracy, and F1-score. Interpretability was enhanced using integrated gradients to identify key predictors.

Results

For the static 12-hour to 1-day mortality prediction task, the TBAL model achieved AUROCs of 95.9 (95% CI 94.2-97.5) and 93.3 (95% CI 91.5-95.3) and AUPRCs of 48.5 and 21.6 in MIMIC-IV and eICU-CRD, respectively. Accuracy and F1-scores reached 94.1 and 46.7 in MIMIC-IV and 92.2 and 28.1 in eICU-CRD. In dynamic prediction tasks, AUROCs reached 93.6 (95% CI 93.2-93.9) and 91.9 (95% CI 91.6-92.1), with AUPRCs of 41.3 and 50, respectively. The model maintained high recall for positive cases (82.6% and 79.1% in MIMIC-IV and eICU-CRD). Cross-database validation yielded AUROCs of 81.3 and 76.1, confirming generalizability. Subgroup analysis showed stable performance across age, sex, and severity strata, with top predictors including lactate, vasopressor use, and Glasgow Coma Scale score.

Conclusions

The TBAL model offers a robust, interpretable, and generalizable solution for dynamic real-time mortality risk prediction in ICU patients. Its ability to adapt to irregular temporal patterns and to provide hourly updated predictions positions it as a promising decision-support tool. Future work should validate its utility in prospective clinical trials and investigate its integration into real-world ICU workflows to enhance patient outcomes.

Subjects

MeSH terms (MEDLINE)

Intensive Care UnitsHospital MortalityHumansRetrospective StudiesElectronic Health RecordsRisk AssessmentFemaleMaleMiddle AgedMachine LearningAgedLongitudinal Studies

Author keywords

continuous predictionin-hospital mortalityintensive care unitsmachine learningmodel interpretability

Authors and institutions

  • Department of Anesthesiology, West China Hospital of Sichuan University, Chengdu, China.
  • West China Biomedical Big Data Center, West China Hospital of Sichuan University, Chengdu, China.
  • Information Management and Data Center, Beijing Tiantan Hospital, Beijing, China.
  • Center for Computational Systems Medicine, McWilliams School of Biomedical Informatics, The University of Texas Health Science Center at Houston, Houston, TX, United States.
  • Department of Scientific Research, The People's Hospital of Yubei District of Chongqing City, Chongqing, China.

ORCID profiles: Zhuo Zheng, Jiawei Luo, Yingchao Zhu, Lei Du, Lan Lan, Xiaobo Zhou, Xiaoyan Yang, Shixin Huang

Cite

APA

Zheng, Z., Luo, J., Zhu, Y., Du, L., Lan, L., Zhou, X., Yang, X., & Huang, S. (2025). Development and Validation of a Dynamic Real-Time Risk Prediction Model for Intensive Care Units Patients Based on Longitudinal Irregular Data: Multicenter Retrospective Study. Journal of Medical Internet Research, 27, e69293. https://doi.org/10.2196/69293

Vancouver

Zheng Z, Luo J, Zhu Y, Du L, Lan L, Zhou X, et al. Development and Validation of a Dynamic Real-Time Risk Prediction Model for Intensive Care Units Patients Based on Longitudinal Irregular Data: Multicenter Retrospective Study. Journal of Medical Internet Research. 2025;27:e69293. doi:10.2196/69293

BibTeX
@article{Zheng2025,
  author = {Zheng, Zhuo and Luo, Jiawei and Zhu, Yingchao and Du, Lei and Lan, Lan and Zhou, Xiaobo and Yang, Xiaoyan and Huang, Shixin},
  title = {Development and Validation of a Dynamic Real-Time Risk Prediction Model for Intensive Care Units Patients Based on Longitudinal Irregular Data: Multicenter Retrospective Study},
  journal = {Journal of Medical Internet Research},
  year = {2025},
  volume = {27},
  pages = {e69293},
  doi = {10.2196/69293}
}
Start for free →