Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

GRU-D Characterizes Age-Specific Temporal Missingness in MIMIC-IV.

Niklas Giesa1, Mert Akguel1, Sebastian Daniel Boie1

  • 1Institute of Medical Informatics, Charité - Universitätsmedizin Berlin, 10117 Berlin.

Studies in Health Technology and Informatics
|May 17, 2025
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

LLM-Assisted Clinical Data Harmonization: Combining Automated ETL Generation with Semantic Vocabulary Mapping for OMOP CDM.

Studies in health technology and informatics·2026
Same author

KoRob: A Berlin-Brandenburg Research Network for Social Robotics and Health Informatics in Transition Societies.

Studies in health technology and informatics·2026
Same author

[Revisiting Risk Classification in the German Maternity Record: First Results from the FeMaR Study].

Zeitschrift fur Geburtshilfe und Neonatologie·2026
Same author

Digital technologies for healthy longevity: A policy agenda from the Einstein circle on inclusive, data-driven ageing societies.

Ageing research reviews·2026
Same author

Calibration of Self-Reported Confidence and Accuracy of Large Language Models in Medical Question Answering.

Journal of medical systems·2026
Same author

Associating liver partition and portal vein ligation for staged hepatectomy (ALPPS) <i>vs.</i> liver transplantation for borderline-resectable colorectal liver metastases: evidence not yet definitive.

Hepatobiliary surgery and nutrition·2026
Same journal

A GenAI Pipeline for Violinist Kinematic Data Management.

Studies in health technology and informatics·2026
Same journal

AMAL-For-Qatar: A Comprehensive AI Ecosystem for Fetal Ultrasound Analysis - Project Overview and Achievements.

Studies in health technology and informatics·2026
Same journal

Longitudinal Treatment-Aware Multimodal AI for Dermatology: A Scoping Review.

Studies in health technology and informatics·2026
Same journal

Predicting Postpartum Depression Using Imbalance-Aware Machine Learning.

Studies in health technology and informatics·2026
Same journal

Validation of Deep-Learning Models for Autosegmentation of Brain Metastases.

Studies in health technology and informatics·2026
Same journal

Delay-Dependent Gating in Modular RNNs.

Studies in health technology and informatics·2026
See all related articles

This study introduces GRU-D, a novel machine learning model, to analyze missing patient data patterns. GRU-D effectively distinguishes between elderly and young patients using vital sign time series, highlighting its potential for advanced imputation techniques.

Area of Science:

  • Clinical Machine Learning
  • Time Series Analysis
  • Healthcare Informatics

Background:

  • Temporal missingness in patient data is an emerging challenge in clinical machine learning.
  • Understanding these unobserved patterns holds significant predictive potential for patient outcomes.

Purpose of the Study:

  • To develop and evaluate a novel deep learning model, GRU-D, for analyzing temporal missingness in clinical time series data.
  • To assess the model's ability to perform binary classification between elderly and young patients based on vital sign data.

Main Methods:

  • Utilized a gated recurrent unit with decay mechanisms (GRU-D) for time series analysis.
  • Input data comprised the first 24 hours of 5 vital signs from the MIMIC-IV database.
  • Model performance was evaluated using Area Under the Receiver Operating Characteristic Curve (AUROC) and Area Under the Precision-Recall Curve (AUPRC).
Keywords:
GRU-DICUMIMIC-IVTemporal Missingness

Related Experiment Videos

Main Results:

  • GRU-D achieved an AUROC of 0.778 and an AUPRC of 0.797 on bootstrapped data.
  • Analysis of model parameters revealed distinct patterns in the temporal missingness of blood pressure and respiratory rate.
  • The model successfully identified differences in data missingness between patient groups.

Conclusions:

  • GRU-D demonstrates efficacy in classifying patients based on vital sign data, considering temporal missingness.
  • The model's ability to interpret missingness patterns provides insights into data characteristics.
  • This work lays the foundation for developing advanced imputation techniques in clinical machine learning.