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 Video

Updated: Sep 16, 2025

Controlled Cortical Impact Model for Traumatic Brain Injury
05:30

Controlled Cortical Impact Model for Traumatic Brain Injury

Published on: August 5, 2014

28.9K

Benchmarking Ensemble Models to Predict Prolonged Hospital Stay in Traumatic Brain Injury: A Single-Institution

Shrinit Babel1, Jade Vanderpool1, Maurice Inkel2

  • 1University Of South Florida Morsani College Of Medicine, Tampa, Florida.

The Journal of Surgical Research
|July 6, 2025
PubMed
Summary

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

Hyperbaric Oxygen Treatment for Enhancing Athletic Recovery: A Systematic Review and Meta-Analysis.

Undersea & hyperbaric medicine : journal of the Undersea and Hyperbaric Medical Society, Inc·2026
Same author

Spatial frequency processing preferentially recruits distributed cortical interactions in V1.

Neuroreport·2026
Same author

Care coordination and procedural outcomes of percutaneous endoscopic gastrostomy and tracheostomy placement in severe traumatic brain injury.

Current problems in surgery·2026
Same author

Acute Disc Herniation at the Proximal Junctional Segment (UIV/UIV + 1) Following Long Segment Spinal Fusion With Cement-Augmented Instrumentation: A Retrospective Study.

International journal of spine surgery·2026
Same author

Evidence-based, cost-effective management of lower gastrointestinal bleeding. An algorithm of the journal of trauma and acute care surgery emergency general surgery algorithms work group.

The journal of trauma and acute care surgery·2026
Same author

Evidence-based, cost-effective management of upper gastrointestinal hemorrhage. An algorithm of the Journal of Trauma and Acute Care Surgery emergency general surgery algorithms work group.

The journal of trauma and acute care surgery·2026
This summary is machine-generated.

Ensemble machine learning models accurately predict prolonged length of stay (PLOS) in traumatic brain injury (TBI) patients. Combining algorithms like XGBoost and ANN improves prediction robustness, outperforming individual models.

Area of Science:

  • Medical Informatics
  • Machine Learning in Healthcare
  • Trauma Surgery

Background:

  • Prolonged length of stay (PLOS) is a key indicator of care quality and efficiency, particularly for traumatic brain injury (TBI) patients.
  • Machine learning (ML) models can predict PLOS, but challenges like class imbalance and limited generalizability hinder real-world application.
  • This study addresses ML model benchmarking and ensemble techniques for PLOS prediction in TBI, considering surgical domain adaptation.

Purpose of the Study:

  • To benchmark existing machine learning models for predicting PLOS in TBI patients.
  • To explore the efficacy of ensemble models in improving PLOS prediction accuracy and robustness.
  • To investigate domain adaptation challenges in surgical settings for TBI PLOS prediction.

Main Methods:

Keywords:
Ensemble learningLength of stayMachine learningNeural networksPredictionTraumatic brain injuryXGBoost

More Related Videos

Detecting Behavioral Deficits in Rats After Traumatic Brain Injury
07:54

Detecting Behavioral Deficits in Rats After Traumatic Brain Injury

Published on: January 30, 2018

18.2K
A Bedside, Single Burr Hole Approach to Multimodality Monitoring in Severe Brain Injury
06:18

A Bedside, Single Burr Hole Approach to Multimodality Monitoring in Severe Brain Injury

Published on: March 26, 2019

9.1K

Related Experiment Videos

Last Updated: Sep 16, 2025

Controlled Cortical Impact Model for Traumatic Brain Injury
05:30

Controlled Cortical Impact Model for Traumatic Brain Injury

Published on: August 5, 2014

28.9K
Detecting Behavioral Deficits in Rats After Traumatic Brain Injury
07:54

Detecting Behavioral Deficits in Rats After Traumatic Brain Injury

Published on: January 30, 2018

18.2K
A Bedside, Single Burr Hole Approach to Multimodality Monitoring in Severe Brain Injury
06:18

A Bedside, Single Burr Hole Approach to Multimodality Monitoring in Severe Brain Injury

Published on: March 26, 2019

9.1K
  • Utilized an anonymized dataset of 263 adult TBI patients from a level 1 trauma ICU.
  • Employed Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), and Artificial Neural Network (ANN) algorithms.
  • Developed ensemble models using the SA2DELA framework and optimized hyperparameters with GridSearchCV.

Main Results:

  • Ensemble models (XGBoost, ANN, SVM) achieved an area under the curve of 0.87, outperforming individual base models.
  • Bias-variance-diversity decomposition revealed complementary strengths between XGBoost and ANN, with SVM providing incremental benefits.
  • Key predictors for PLOS included age, body mass index, and injury severity score.

Conclusions:

  • This study pioneers the benchmarking of ML models and the use of ensemble techniques for TBI PLOS prediction.
  • Ensemble methods enhance model robustness in data-limited and varied clinical settings.
  • Future research should explore multiclass/regression models and advanced domain adaptation strategies.