Please use this identifier to cite or link to this item:
Author(s): Sk. Saddam Ahmed
Nilanjan Dey
Amira S. Ashour
Dimitra Sifaki-Pistolla
Dana Balas-Timar
Valentina E. Balas
João Manuel R. S. Tavares
Title: Effect of fuzzy partitioning in Crohn's disease classification: a neuro-fuzzy-based approach
Issue Date: 2017-01
Abstract: Crohn's disease (CD) diagnosis is a tremendouslyserious health problem due to its ultimately effecton the gastrointestinal tract that leads to the need of complexmedical assistance. In this study, the backpropagationneural network fuzzy classifier and a neuro-fuzzy modelare combined for diagnosing the CD. Factor analysis isused for data dimension reduction. The effect on the systemperformance has been investigated when using fuzzypartitioning and dimension reduction. Additionally, furthercomparison is done between the different levels of thefuzzy partition to reach the optimal performance accuracylevel. The performance evaluation of the proposed systemis estimated using the classification accuracy and othermetrics. The experimental results revealed that the classificationwith level-8 partitioning provides a classificationaccuracy of 97.67 %, with a sensitivity and specificity of96.07 and 100 %, respectively.
Subject: Ciências Tecnológicas, Ciências médicas e da saúde
Technological sciences, Medical and Health sciences
Scientific areas: Ciências médicas e da saúde
Medical and Health sciences
Document Type: Artigo em Revista Científica Internacional
Rights: openAccess
Appears in Collections:FEUP - Artigo em Revista Científica Internacional

Files in This Item:
File Description SizeFormat 
171381.pdfDraft762.65 kBAdobe PDFThumbnail
171381.1.png1st Page349.03 kBimage/pngThumbnail

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.