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

Neural network and linear regression models in residency selection

S Pilon1, D Tandberg

  • 1Department of Emergency Medicine, University of New Mexico School of Medicine, Albuquerque 87131-5246, USA.

The American Journal of Emergency Medicine
|July 1, 1997
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

Network geometry of evidence from randomised controlled trials addressing donor selection and source of haematopoietic progenitor cells used in allogeneic transplantation: a systematic scoping review.

Transfusion medicine (Oxford, England)·2018
Same author

Costs of the in-home patients affected by dementia.

Archives of gerontology and geriatrics·2009
Same author

Depressive symptoms and quality of life in elderly patients with exacerbation of chronic obstructive pulmonary disease or cardiac heart failure: preliminary data of a randomized controlled trial.

Archives of gerontology and geriatrics·2007
Same author

The relationship of sleep quality and posttraumatic stress to potential sleep disorders in sexual assault survivors with nightmares, insomnia, and PTSD.

Journal of traumatic stress·2002
Same author

Imagery rehearsal therapy for chronic nightmares in sexual assault survivors with posttraumatic stress disorder: a randomized controlled trial.

JAMA·2001
Same author

Treatment of chronic nightmares in adjudicated adolescent girls in a residential facility.

The Journal of adolescent health : official publication of the Society for Adolescent Medicine·2001
Same journal

Diagnostic accuracy of emergency department triage systems for predicting clinical severity: A systematic review and meta-analysis of five-level triage scales.

The American journal of emergency medicine·2026
Same journal

Methanol toxicity from inhalational abuse of a methyl acetate-containing nail polish remover.

The American journal of emergency medicine·2026
Same journal

Regarding strategies, feasibility of implementations and results of HIV screening in emergency departments.

The American journal of emergency medicine·2026
Same journal

Rocuronium dosing for rapid sequence intubation: A retrospective analysis in ED and ICU settings.

The American journal of emergency medicine·2026
Same journal

Clinical features of adults with undiagnosed acute leukemia in the emergency department: A descriptive study.

The American journal of emergency medicine·2026
Same journal

Occult intraperitoneal gynecological hemorrhage: An under-evaluated source of hemoperitoneum after blunt trauma.

The American journal of emergency medicine·2026
See all related articles

Artificial neural networks and linear regression models equally predict residency applicant rankings. Both methods aid admissions committees in creating final rank lists for the National Residency Match Program (NRMP).

Area of Science:

  • Medical Education
  • Machine Learning in Healthcare
  • Residency Admissions

Background:

  • Multiple linear regression models have historically guided residency program applicant ranking.
  • These preliminary lists assist admissions committees in finalizing rankings for the National Residency Match Program (NRMP).

Purpose of the Study:

  • To compare the predictive accuracy of artificial neural networks against traditional linear regression models for residency applicant ranking.

Main Methods:

  • A prospective cohort study evaluated 74 emergency medicine residency applicants.
  • Data included grades, autobiography, interviews, recommendations, and National Board scores, normalized to account for interviewer variability.
  • Multivariate linear regression and neural network models were trained on 5 years of historical applicant data.

Related Experiment Videos

Main Results:

  • The neural network model achieved a correlation coefficient of 0.77 and R2 of 59.4%.
  • The linear regression model showed a correlation coefficient of 0.74 and R2 of 54.0%.
  • No statistically significant difference was found between the predictive performance of the two models (chi 2 = 1.08, P = .7).

Conclusions:

  • Artificial neural networks perform comparably to linear regression models in forecasting residency applicant rank order.
  • Both methods can serve as valuable tools for residency admissions committees, aiding in the development of final rank lists.