UnissResearch

Logo Universitàegli studi di Sassari
titoli, abstracts, parole chiave >>>
Classifiers trained on dissimilarity representation of medical pattern: a comparative study

Masala, Giovanni Luca Christian and Golosio, Bruno and Oliva, Piernicola and Cascio, Donato and Fauci, Francesco and Tangaro, Sonia and Quarta, Maurizio and Cheran, Sorin Cristian and López Torres, Ernesto (2005) Classifiers trained on dissimilarity representation of medical pattern: a comparative study. Il Nuovo cimento. C, Vol. 28 (6), p. 905-912. eISSN 1826-9885. Article.

Full text not available from this repository.

DOI: 10.1393/ncc/i2005-10162-9

Abstract

In this paper we investigate the feasibility of some typical techniques of pattern recognition for the classification of medical examples. The learning of the classifiers is not made in the traditional features space but it can be made by constructing decision rules on dissimilarity (distance) representations. In such a recognition process a new object is described by its distances to (a subset of) the training samples. Purpose of this work is the development of an automatic classification system which could be useful for radiologists in the investigation of breast cancer. The software has been designed in the framework of the MAGIC-5 collaboration. In the automatic classification system the suspicious regions with high probability to include a lesion are extracted from the image as regions of interest (ROIs). Each ROI is characterized by some features extracted from co-occurrence matrix containing spatial statistics information on ROI pixel gray tones. A dissimilarity representation of these features is made before the classification. A Feed-Forward Neural Network (FF-NN), a K-Nearest Neighbour (K-NN) and a Linear Discriminant Analysis (LDA) are employed to distinguish pathological records from not-pathological ones by the new features. The results obtained in terms of sensitivity (percentage of pathological ROIs correctly classified) and specificity (percentage of healthy ROIs correctly classified) will be comparatively presented. The K-NN classifier gives slightly better results than FF-NN and LDA accuracy (percentage of cases correctly classified) on two-classes problem (pathologic or healthy patients).

Item Type:Article
ID Code:1907
Status:Published
Refereed:Yes
Uncontrolled Keywords:Feed-Forward Neural Network (FF-NN), K-Nearest Neighbour (K-NN), Linear Discriminant Analysis (LDA), breast cancer, regions of interest (ROIs)
Subjects:Area 02 - Scienze fisiche > FIS/07 Fisica applicata (a beni culturali, ambientali, biologia e medicina)
Divisions:001 Università di Sassari > 03 Istituti > Matematica e fisica
Publisher:Società Italiana di Fisica
eISSN:1826-9885
Deposited On:18 Aug 2009 10:06

I documenti depositati in UnissResearch sono protetti dalle leggi che regolano il diritto d'autore

Repository Staff Only: item control page