Author : Miss. Ankita Satyendra Singh 1
Date of Publication :21st June 2017
Abstract: High False Negative Rate (FNR) is a very significant problem in a Computer Aided Diagnostic System as false negative answer may lead to a very high increase in the number of deaths. The main aim of this paper lies in the development of a new Computer Aided Diagnosis (CADx) system for the proper identification of breast masses. It also focuses at extraction of textural features. The input images are pre-processed by using Adaptive Median Filter and then segmented by using Gaussian Mixture Model i.e. GMM segmentation and further are subjected to feature extraction, selection and finally classification by using PNN classifier. MIAS database is used for research purpose which contains 322 mammogram images out of which 60 images as 20 of benign, 20 malignant and 20 normal are taken into consideration for feature extraction. 22 texture features are extracted and are further classified. PNN classifier with 80-20 train-test partition is used for classification. The Sensitivity, Specificity and Accuracy obtained by the selected features are 100%, 100%, and 100% respectively.
Reference :
-
1) http://www.babymed.com/cancer/introduction-breastcancer-causes-and-risk-factors
3) https://www.researchgate.net/publication/23081760 _Basics_of_Oncology
4) http://www.wcrf.org/int/cancer-facts-figures/data specific-cancers/breast-cancer-statistics
5) Bartosz Krawczyka, Gerald Schaefer “A hybrid classifier committee for analyzing asymmetry features in breast thermograms” Applied Soft Computing 20 (2014) 112–118
6) Marcelo Zanchetta do Nascimento, Alessandro Santana Martins, Leandro Alves Neves, Rodrigo Pereira Ramos, Edna Lucia Flores, Gilberto Arantes Carrijo “Classification of masses in mammographic image using wavelet domain features and polynomial classifier ”Expert Systems with Applications 40 (2013) 6213–6221.
7) Danilo Cesar Pereiraa, Rodrigo Pereira Ramosb, Marcelo Zanchetta do Nascimento “Segmentation and detection of breast cancer in mammograms combining wavelet analysis and genetic algorithm” Computer methods