Computerized Consensus Diagnosis: A Classification Strategy for the Robust Analysis of MR spectra. I. Application to 1H Spectra of Thyroid Neoplasms

Created by W.Langdon from gp-bibliography.bib Revision:1.4202

  author =       "Ray L. Somorjai and Alexander E. Nikulin and 
                 Nic Pizzi and Dick Jackson and Gordon Scarth and 
                 Brion Dolenko and Heather Gordon and Peter Russell and 
                 Cynthia L. Lean and Leigh Delbridge and Carolyn E. Mountford and 
                 Ian C. P. Smith",
  title =        "Computerized Consensus Diagnosis: A Classification
                 Strategy for the Robust Analysis of {MR} spectra. {I}.
                 Application to {1H} Spectra of Thyroid Neoplasms",
  journal =      "Magnetic Resonance Medicine",
  year =         "1995",
  volume =       "33",
  number =       "2",
  pages =        "257--263",
  month =        feb,
  keywords =     "genetic algorithms, genetic programming, computerised
                 consensus diagnosis, robust classification, thyroid
                 neoplasms, proton magnetic resonance spectrum, LDA,
                 ANN, GEPPETTO",
  ISSN =         "0740-3194",
  URL =          "",
  URL =          "",
  DOI =          "doi:10.1002/mrm.1910330217",
  size =         "7 pages",
  abstract =     "We introduce and apply a new classification strategy
                 we call computerised consensus diagnosis (CCD). Its
                 purpose is to provide robust, reliable classification
                 of biomedical data. The strategy involves the
                 cross-validated training of several classifiers of
                 diverse conceptual and methodological origin on the
                 same data, and appropriately combining their outcomes.
                 The strategy is tested on proton magnetic resonance
                 spectra of human thyroid biopsies, which are
                 successfully allocated to normal or carcinoma classes.
                 We used Linear Discriminant Analysis, a Neural
                 Net-based method, and Genetic Programming as
                 independent classifiers on two spectral regions, and
                 chose the median of the six classification outcomes as
                 the consensus. This procedure yielded 100percent
                 specificity and 100percent sensitivity on the training
                 sets, and 100percent specificity and 98percent
                 sensitivity on samples of known malignancy in the test
                 sets. We discuss the necessary steps any classification
                 approach must take to guarantee reliability, and stress
                 the importance of fuzziness and undecidability in
                 robust classification.",
  notes =        "PMID: 7707918 [PubMed - indexed for MEDLINE] consensus
                 means taking 2 of 3 vote from the three different
                 classifiers (GP, LDA and NN). PCA",

Genetic Programming entries for Ray L Somorjai Alexander E Nikulin Nicolino J Pizzi Dick Jackson Gordon Scarth Brion Dolenko Heather Gordon Peter Russell Cynthia L Lean Leigh Delbridge Carolyn E Mountford Ian C P Smith