Application of multivariate, fuzzy set and neural network analysis in quantitative cytological examinations

Bela Molnar, Zoltan Szentirmay, Miklos Bodo, Janos Sugar, Janos Feher

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

Multivariate statistical methods have been used in several studies to increase the diagnostic reliability of TV image analyser systems. In recent years some algorithms for decision support (fuzzy logic) and for pattern recognition (neural nets), both non-linear, were developed. This paper reports on preliminary results obtained with these methods in quantitative cytology and compares them to the traditional classifiers. A total of 21 normal, 15 dysplastic and 23 malignant, gastric imprint smears were Feulgen stained and analysed on a Leitz Miamed DNA cytophotometer system. Mean DNA content, the 2c deviation index (2cDI), 5c exceeding rate (5cER), G1, S, G2 phase fraction ratios, cell nucleus area and form factor were determined. Diagnostic accuracy of the discriminant analysis was 96% for the malignant cases, 87% for dysplasias and 81% for normal cases. Cluster analysis gave no significant result. Our diagnostic system utilizing fuzzy logic has made the diagnostic borders adjustable and reliable. The back-propagation neural net correctly classifed the normal and malignant cases (100%) and all but one of the dysplasias (98%). The non-linear mathematical methods improved the reliability of the diagnostic system. These new algorithms gave results comparable to traditional classifiers. The application of these methods to clinical samples is encouraging.

Original languageEnglish
Pages (from-to)161-175
Number of pages15
JournalAnalytical Cellular Pathology
Volume5
Issue number3
Publication statusPublished - May 1 1993

Keywords

  • Back propagation
  • Cluster analysis
  • Discriminant analysis
  • Fuzzy logic
  • Gastric tumours
  • Neural net
  • TV cytometry

ASJC Scopus subject areas

  • Pathology and Forensic Medicine
  • Cell Biology

Fingerprint Dive into the research topics of 'Application of multivariate, fuzzy set and neural network analysis in quantitative cytological examinations'. Together they form a unique fingerprint.

  • Cite this