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

Classification of crossed immunoelectrophoretic patterns using digital image processing and artificial neural

I Søndergaard1, B N Krath, M Hagerup

  • 1Chemistry Department, Royal Veterinary and Agricultural University, Frederiksberg, Denmark.

Electrophoresis
|July 1, 1992
PubMed
Summary

This study introduces a novel method for artificial neural networks to analyze crossed immunoelectrophoresis patterns. Adding noise during training significantly improved pattern recognition and classification accuracy.

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

Non-haem iron availability from pork meat: Impact of heat treatments and meat protein dose.

Meat science·2011
Same author

The T cell response to major grass allergens is regulated and includes IL-10 production in atopic but not in non-atopic subjects.

International archives of allergy and immunology·2010
Same author

Exploring abiotic stress on asynchronous protein metabolism in single kernels of wheat studied by NMR spectroscopy and chemometrics.

Journal of experimental botany·2009
Same author

A wake-up call: are degenerative diseases provoked by some of our normal food constituents? A case for nutrigenomics.

Medical hypotheses·2006
Same author

Induction of systemic CTL responses in melanoma patients by dendritic cell vaccination: cessation of CTL responses is associated with disease progression.

International journal of cancer·2001
Same author

Implications of secondary structure prediction and amino acid sequence comparison of class I and class II phosphoribosyl diphosphate synthases on catalysis, regulation, and quaternary structure.

Protein science : a publication of the Protein Society·2001

Area of Science:

  • Biotechnology
  • Artificial Intelligence
  • Immunology

Background:

  • Crossed immunoelectrophoresis (CIE) is a powerful technique for analyzing complex protein mixtures.
  • Interpreting CIE patterns can be subjective and time-consuming.
  • Automating CIE pattern analysis holds significant potential for research and diagnostics.

Purpose of the Study:

  • To develop a method for presenting CIE patterns to artificial neural networks (ANNs).
  • To train an ANN to accurately classify CIE patterns.
  • To enhance the generalization ability and efficiency of ANNs in analyzing biological data.

Main Methods:

  • Representing CIE patterns as three-dimensional vectors for ANN input.
  • Training ANNs on these vector representations.

Related Experiment Videos

  • Incorporating noise injection into the training data to improve model robustness.

    • Main Results:

    • The ANN successfully learned to classify CIE patterns presented as 3D vectors.
    • Noise addition during training substantially increased the ANN's ability to generalize.
    • Noise injection reduced the number of training iterations required to achieve a target error level.
    • The trained ANN achieved 1% error in classifying distorted CIE patterns.

    Conclusions:

    • ANNs can be effectively trained to interpret complex CIE patterns.
    • Noise injection is a valuable technique for improving ANN performance in pattern recognition tasks.
    • This method offers a promising approach for automated analysis of immunoelectrophoretic data.