Detection of Malignant Tumour in Mammography Images Using Artificial Neural Networks with Fuzzy Rules

Abstract

Breast cancer is a collection of cancer cells that starts in the breast cells and it expands from tissue of breast. Now a day Mammogram is one technique to detect the breast cancer earlyusing x-ray image of breast and it is used to reduce the deaths of breast cancer. This breast cancer disease is curable if discovered starting stage. This paper studies different methods utilized for the detection of breast cancer using mammogram classification. In this paper, the feature extraction and classification of mammogram image can be done by the artificial neural networks. Different kinds of feature extraction from mammogram image to detecting the bread cancer contains shape, position and surface features etc., this image feature extraction is significant in classification of image. By utilizing the image processing these image features are extracted. Image segmentation is performed for feature extraction of mammogram image, in this process image is partitioned into multiple segments, therefore when change the image representation into something that is more significant and simple to examine. Here the fuzzy rules are introduced to process the related data from cases of breast cancer in mammogram image in order to give the risk diagnosis of breast cancer. The preprocessing method is used to sustain an effectiveness of image by correct and adjusting the mammogram image and also it is used to improve the image quality and create it ready for additional working by reducing the unrelated noise to provide new brightness value in output image it is called as filtration and unwanted parts of background of mammogram image is eliminated. Some techniques are discussed for mammogram image classification to earlier detection of breast cancer.