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

VALAB: expert system for validation of biochemical data.

P M Valdiguié1, E Rogari, H Philippe

  • 1Laboratoire de Biochimie, Centre Hospitalier Universitaire de Rangueil, Toulouse, France.

Clinical Chemistry
|January 1, 1992
PubMed
Summary

Artificial intelligence (AI) systems like VALAB can validate laboratory data in real time. While VALAB showed high sensitivity in detecting abnormal results, human experts demonstrated superior specificity.

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

[Acardiac twins: pronostics markers' study].

Gynecologie, obstetrique & fertilite·2011
Same author

Improved phylogenomic taxon sampling noticeably affects nonbilaterian relationships.

Molecular biology and evolution·2010
Same author

[European normative recommendations for medical gas pipeline systems].

Annales francaises d'anesthesie et de reanimation·2005
Same author

Sure facts and open questions about the origin and evolution of photosynthetic plastids.

Research in microbiology·2002
Same author

Heterotachy, an important process of protein evolution.

Molecular biology and evolution·2001
Same author

The last universal common ancestor (LUCA), simple or complex?

The Biological bulletin·2001

Area of Science:

  • Clinical Chemistry
  • Medical Informatics
  • Artificial Intelligence

Background:

  • High-throughput laboratories generate vast amounts of data requiring efficient validation.
  • Artificial intelligence (AI) offers potential solutions for automating complex decision-making processes in clinical settings.

Purpose of the Study:

  • To describe the development and performance of VALAB, an AI-powered expert system for real-time laboratory data validation.
  • To compare the diagnostic accuracy of VALAB against human laboratory experts.

Main Methods:

  • VALAB, an expert system, was developed using over 4000 rules.
  • The system analyzes data based on result correlations, physiological associations, test origin, and patient demographics (age, sex).
  • Performance was evaluated using 200 randomly selected abnormal chemistry profiles.

Related Experiment Videos

Main Results:

  • VALAB achieved a sensitivity of 0.75 in detecting abnormal cases, surpassed by only one of seven human experts.
  • Human experts demonstrated higher specificity (0.63 for VALAB) in validating results.
  • The VALAB system has validated over 50,000 patient reports since its implementation in November 1988.

Conclusions:

  • AI expert systems like VALAB can effectively assist in real-time laboratory data validation.
  • While AI shows promise, human expertise remains crucial for achieving optimal diagnostic specificity.
  • VALAB represents a significant advancement in automating data validation in large-scale laboratory operations.