Architectural Distortion Detection in Mammogram using Contourlet Transform and Texture Features

Abstract

Breast Cancer is one of the most affecting diseases in the world. Architectural Distortion is one of the indications of breast cancer. It is an abnormal arrangement of tissue strands of the breast, often a radial or perhaps a somewhat random pattern, but without any associated mass as the apparent cause of this distortion. This project used Contourlet Transform based method to detect the location of the architectural distortion in mammograms. The dominant angle is detected in this Contourlet decomposition which comprises of Laplacian Pyramidal Filter bank and Steerable Gaussian Filter. The textural features will not show any deviation in the distorted area. Here the methods proposed will assist the Radiologist to identify the unidentified potential sites of distortion. Therefore the main aim of this work becomes the reduction of the region of interests. In this project 39 cases were testes in which 19 are Architectural Distorted image from the MIAS database and 17 Normal images. The proposed algorithm detected 1502 regions of interest (ROIs) with 247 True Positives and 1255 False Positives. In this paper 26 textural features were considered and result is obtained as 0.77 with an artificial neural network. The proposed methods have significant result in detecting the architectural distortion in mammograms of interval cancer cases.

Segmentation of brain tumour using Enhanced Thresholding Algorithm and Calculatethe area of the tumour

“…The kernel of the high pass filter is designed to increase the brightness of the center pixel relative to neighboring pixels. [10] …”

“…The pattern of neighbors is called the "window", which slides, entry by entry, over the entire signal. [10][11] This filter enhance the quality of the MRI image.…”

Detection and Classification of Breast Cancer Using Support Vector Machine and Artificial Neural Network Using Contourlet Transform

“…The Contourlet transformation (CT) technique used in this work, is a two dimensional transformation technique which contain properties like multi-resolution and directionality. Both spatial and frequency domain features are extracted as feature values using GLCM method [11] [13].…”

A New Method for Detecting Architectural Distortion in Mammograms by NonSubsampled Contourlet Transform and Improved PCNN

“…Biswas et al proposed a Gaussian Mixture Model (GMM) to detect AD with a sensitivity of 84% [10]. Anand et al proposed using contourlet transform and neural network to detect AD, with an accuracy of 64% [11]. Rangayyan et al proposed employing the angular deviation degree of the linear structure of breast tissue as a feature to detect AD [12].…”