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 Concept Videos

Strategies for Assessing and Addressing Confounding01:25

Strategies for Assessing and Addressing Confounding

93
Confounding is a critical issue in epidemiological studies, often leading to misleading conclusions about associations between exposures and outcomes. It occurs when the relationship between the exposure and the outcome is mixed with the effects of other factors that influence the outcome. Given that, addressing confounding is of high importance for drawing accurate inferences in research.
Confounding can be addressed at both the design phase of a study and through analytical methods after data...
93
Regression Toward the Mean01:52

Regression Toward the Mean

6.3K
Regression toward the mean (“RTM”) is a phenomenon in which extremely high or low values—for example, and individual’s blood pressure at a particular moment—appear closer to a group’s average upon remeasuring. Although this statistical peculiarity is the result of random error and chance, it has been problematic across various medical, scientific, financial and psychological applications. In particular, RTM, if not taken into account, can interfere when...
6.3K

You might also read

Related Articles

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

Sort by
Same author

Sex-specific effects of manual pressure augmentation during defibrillation: a mechanistic hypothesis.

Resuscitation·2026
Same author

Improving surgical team confidence for intraoperative cardiac arrest in non-standard positions (sitting beach chair and prone): a prospective in-situ simulation study with three-month follow-up.

Resuscitation plus·2026
Same author

Prognostic performance of plasma and urinary biomarkers of kidney injury and function after out-of-hospital cardiac arrest.

Resuscitation·2026
Same author

Cardiometabolic multimorbidity and survival after out-of-hospital cardiac arrest.

Resuscitation plus·2026
Same author

Delayed awakening after out-of-hospital cardiac arrest: a scoping review of definitions, determinants, and prognostic implications.

Resuscitation·2026
Same author

Reperfusion treatment strategies in acute pulmonary embolism: Study protocol for a multicenter prospective observational study in the Nordic countries (PE-NORDIC).

Thrombosis research·2026

Related Experiment Video

Updated: Jun 27, 2025

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
12:18

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

7.5K

Improved interpretable machine learning emergency department triage tool addressing class imbalance.

Clarisse Sj Look1, Salinelat Teixayavong1, Therese Djärv2

  • 1Health Services and Systems Research, Duke-NUS Medical School, Singapore, Singapore.

Digital Health
|May 6, 2024
PubMed
Summary

Addressing class imbalance in the Score for Emergency Risk Prediction (SERP) improved its performance. The new SERP+ model enhances emergency department triage accuracy by better stratifying patient risk.

Keywords:
Machine learningemergency departmentinterpretable‌triage

More Related Videos

Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index
06:55

Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index

Published on: January 8, 2020

14.5K
Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.2K

Related Experiment Videos

Last Updated: Jun 27, 2025

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
12:18

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

7.5K
Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index
06:55

Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index

Published on: January 8, 2020

14.5K
Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.2K

Area of Science:

  • Machine Learning in Healthcare
  • Clinical Triage and Risk Prediction
  • Emergency Medicine Analytics

Background:

  • The Score for Emergency Risk Prediction (SERP) is a machine learning tool designed to aid emergency department (ED) triage decisions.
  • Initial SERP models demonstrated good predictive performance for short-term mortality but were developed on a dataset with significant class imbalance.
  • Class imbalance in medical datasets can potentially compromise the predictive accuracy of machine learning models.

Purpose of the Study:

  • To investigate whether addressing class imbalance during the development of SERP can enhance its predictive performance.
  • To determine if an improved SERP model (SERP+) can lead to more accurate triage decisions in the ED.
  • To compare the performance of the modified SERP+ scores against the original SERP and other standard triage risk scores.

Main Methods:

  • Utilized a large dataset from Singapore General Hospital's ED (1,833,908 records, 2008-2020).
  • Employed random oversampling and undersampling techniques within the AutoScore-Imbalance framework to develop SERP+ scores.
  • Compared the predictive performance (AUC, sensitivity, specificity, etc.) of SERP+ against SERP and common triage scores on separate test sets.

Main Results:

  • The developed SERP+ scores, comprising five to six variables, showed higher AUC values (0.874-0.905) compared to original SERP scores (0.859-0.894).
  • Statistical significance was observed for SERP+-7d and SERP+-30d, indicating superior predictive power after addressing class imbalance.
  • SERP+ demonstrated marginal improvements in sensitivity, specificity, balanced accuracy, and positive predictive value, outperforming common triage risk scores.

Conclusions:

  • Addressing class imbalance during the training phase significantly improved the performance of the SERP risk prediction scores.
  • The enhanced SERP+ model offers better patient risk stratification, which is crucial for effective ED triage.
  • Machine learning-based scores like SERP+ hold substantial potential for supporting accurate, data-driven triage decisions in emergency departments.