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

Predicting active pulmonary tuberculosis using an artificial neural network.

A A El-Solh1, C B Hsiao, S Goodnough

  • 1Department of Medicine, Erie County Medical Center, Buffalo, NY 14215, USA. solh@buffalo.edu

Chest
|October 26, 1999
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

Applications of airway pressure release ventilation.

Lung·2010
Same author

Noninvasive ventilation for prevention of post-extubation respiratory failure in obese patients.

The European respiratory journal·2006
Same author

Clinical factors associated with hyperkalemia in patients with congestive heart failure.

Journal of clinical pharmacy and therapeutics·2005
Same author

The utility of neural network in the diagnosis of Cheyne-Stokes respiration.

Journal of medical engineering & technology·2003
Same author

Epidemiologic evidence linking antioxidant vitamins to pulmonary function and airway obstruction.

Epidemiologic reviews·2002
Same author

Outcome of older patients with severe pneumonia predicted by recursive partitioning.

Journal of the American Geriatrics Society·2002
Same journal

The effect of a two-day stay at high altitude (2500 m) on right ventricular afterload and oxygen delivery in patients with pulmonary vascular disease A randomized controlled crossover trial.

Chest·2026
Same journal

A Comparative Study of Radiation Exposure in Conventional and Robotic Bronchoscopy.

Chest·2026
Same journal

Independent Prognostic Contributions of Anti-Ro52 and Anti-MDA5 in Autoimmune-Associated Interstitial Lung Disease.

Chest·2026
Same journal

Lung aeration and gas exchange in SGA or AGA infants with moderate-severe BPD: secondary analysis of the PATH-BPD study.

Chest·2026
Same journal

Lung Cancer Incidence and Mortality after Negative Low-Dose CT Screening Results.

Chest·2026
Same journal

Symptom prevalence and impact on lung cancer risk in the SUMMIT study.

Chest·2026
See all related articles

An artificial neural network accurately predicts active pulmonary tuberculosis (TB) in patients, outperforming physician assessments. This AI model is crucial for preventing nosocomial TB outbreaks by identifying infectious cases early.

Area of Science:

  • Medical Informatics
  • Artificial Intelligence in Medicine
  • Infectious Disease Epidemiology

Background:

  • Nosocomial tuberculosis (TB) outbreaks are often linked to undiagnosed pulmonary TB.
  • Accurate identification of active TB index cases is vital for disease transmission control.

Purpose of the Study:

  • To develop an artificial neural network (ANN) for predicting active pulmonary TB.
  • To compare the ANN's predictive accuracy against physician assessments at healthcare facility presentation.

Main Methods:

  • A nonconcurrent prospective study was conducted at a university-affiliated hospital.
  • A general regression neural network (GRNN) model was developed using clinical and radiographic data.
  • The predictive accuracy of the GRNN was evaluated against clinicians' assessments in derivation and validation groups.

Related Experiment Videos

Main Results:

  • The GRNN model demonstrated significantly higher predictive accuracy (c-index: 0.947) than physician predictions (c-index: 0.61) in the derivation group (p < 0.001).
  • In the validation group, the GRNN achieved a c-index of 0.923, compared to 0.716 for physician assessments.
  • The GRNN's performance, measured by the area under the receiver operating characteristic curve (c-index), was consistently superior.

Conclusions:

  • An artificial neural network model can effectively identify patients with active pulmonary TB.
  • The developed ANN surpasses physician clinical assessment in accuracy for diagnosing active pulmonary TB.
  • This AI-driven approach holds potential for improving TB control strategies.