Campos L F A, Silva A C, Barros A K
Laboratory for Biological Information Processing, University Federal of Maranhão, São Luis, Brazil.
Methods Inf Med. 2007;46(2):212-5.
This paper proposes an efficient method for the discrimination and classification of mammograms with benign, malignant and normal tissues.
The proposed method consists of selection of tissues, feature extraction using independent component analysis, feature selection by the forward-selection technique and classification of the tissue by the multilayer perceptron.
The method is tested for a mammogram set of the MIAS database, resulting in a 97.83% success rate, with 98.0% specificity and 97.5% sensitivity.
The proposed method showed a good classification rate. The method will be useful for early cancer diagnosis.
