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

Near-optimal region selection for feature space reduction: novel preprocessing methods for classifying MR spectra

A E Nikulin1, B Dolenko, T Bezabeh

  • 1Institute for Biodiagnostics, National Research Council, Winnipeg, Manitoba, Canada.

NMR in Biomedicine
|August 27, 1998
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

Creating robust, reliable, clinically relevant classifiers from spectroscopic data.

Biophysical reviews·2017
Same author

Feasibility of infrared spectroscopy with pattern recognition techniques to identify a subpopulation of mares at risk of producing foals diagnosed with failure of transfer of passive immunity.

Australian veterinary journal·2012
Same author

Class proximity measures--dissimilarity-based classification and display of high-dimensional data.

Journal of biomedical informatics·2011
Same author

Potential of mid-infrared spectroscopy to aid the triage of patients with acute chest pain.

The Analyst·2009
Same author

Detecting colorectal cancer by 1H magnetic resonance spectroscopy of fecal extracts.

NMR in biomedicine·2009
Same author

Detection of cholangiocarcinoma with magnetic resonance spectroscopy of bile in patients with and without primary sclerosing cholangitis.

Acta radiologica (Stockholm, Sweden : 1987)·2008

A new genetic algorithm-guided method effectively preprocesses biomedical magnetic resonance spectra for disease classification. This approach achieves over 97% accuracy in identifying brain tumors and colorectal cancer stages.

Area of Science:

  • Biomedical Engineering
  • Medical Informatics
  • Computational Biology

Background:

  • Accurate classification of diseases using biomedical magnetic resonance (MR) spectra is crucial for diagnosis and treatment.
  • Preprocessing MR spectra is essential to extract relevant biochemical information for disease identification.
  • Existing methods may not optimally select spectral subregions that retain critical diagnostic information.

Purpose of the Study:

  • To introduce a novel global feature extraction method for preprocessing biomedical MR spectra.
  • To compare the performance of a genetic algorithm-guided method with an enhanced forward selection algorithm.
  • To identify optimal spectral subregions for disease classification.

Main Methods:

  • A genetic algorithm-guided global feature extraction method was developed.

Related Experiment Videos

  • This method was compared against an enhanced standard forward selection algorithm.
  • Both algorithms aimed to select optimal spectral subregions for classification tasks.

    • Main Results:

    • Both the genetic algorithm-guided and enhanced forward selection methods achieved high classification accuracies.
    • Accuracies exceeded 97% in discriminating between meningioma and astrocytoma in brain biopsies.
    • Over 97% accuracy was obtained in classifying colorectal biopsies as normal or tumorous.

    Conclusions:

    • The developed genetic algorithm-guided feature extraction method is effective for biomedical MR spectral preprocessing.
    • Optimal spectral subregion selection significantly aids in disease classification.
    • These methods demonstrate potential for accurate disease diagnosis and staging through MR spectroscopy.