Grading Urinary Bladder Tumors Using Unsupervised Hebbian Algorithm for Fuzzy Cognitive Maps

• PAPAGEORGIOU E. I.

• SPYRIDONOS P. P.

• STYLIOS C. D.

• NIKIFORIDIS G. C.

• GROUMPOS P. P.

Abstract

The application of Fuzzy Cognitive Maps as a modeling and classification tool, for assessing tumors grade for urinary bladder, is examined in this research work. One hundred twenty nine cases were classified according to the WHO grading system in two classes, by experienced pathologists : Low Grade and High Grade, based on eight significant histopathological features that histopathologists selected for each case. This research work incorporates doctor's knowledge in developing the FCM model for tumor grading and utilizes the Nonlinear Hebbian Learning algorithm to further train the FCM and thus to achieve tumor malignancy classification. The classification is based on the histopathological characteristics of tissue that features are the concepts of the Fuzzy Cognitive Map model that was trained using the unsupervised learning algorithm. The classification accuracy is 93.18% for High Grade tumor cases and 90.59%, for tumors of Low Grade.

Journal

• International Journal of Biomedical Soft Computing and Human Sciences: the official journal of the Biomedical Fuzzy Systems Association

  International Journal of Biomedical Soft Computing and Human Sciences: the official journal of the Biomedical Fuzzy Systems Association 9 (2), 33-39, 2003

  Biomedical Fuzzy Systems Association

Keywords

• Fuzzy Cognitive Maps
• Tumor Grading
• Unsupervised Learning
• Nonlinear Hebbian Learning Algorithm
• Urinary Bladder

Details

• CRID: 1390282681057246208
• NII Article ID: 110003963833
• DOI: 10.24466/ijbschs.9.2_33
• ISSN: 2424256X
• ISSN: 21852421
• Text Lang: en
• Data Source:
  • JaLC
  • CiNii Articles
• Abstract License Flag: Disallowed