Feature Extraction of Mammograms

Abstract

Cancer is uncontrolled growth of cells. Breast Cancer is the uncontrolled growth of cells in the breast region. Breast cancer is the second leading cause of cancer deaths in women today. Early detection of the cancer can reduce mortality rate. Early detection of Breast Cancer can be achieved using Digital Mammography, typically through detection of characteristic masses and/or microcalcifications. A Mammogram is an x-ray of the breast tissue which is designed to identify abnormalities. Studies have shown that radiologists can miss the detection of a significant proportion of abnormalities in addition to having high rates of false positives. Therefore, it would be valuable to develop a computer aided method for mass/tumor classification based on extracted features from the Region Of Interest (ROI) in mammograms. ROI has to be segmented from the digital mammogram using the Segmentation techniques. Pattern recognition in image processing requires the extraction of features from regions of the image, and the processing of these features with a pattern recognition algorithm. We consider the feature extraction part of this processing, with a focus on the problem of tumor detection in digital mammography. Features are nothing but observable patterns in the image which gives some information about the image. For every Pattern Classification problem, the most important stage is Feature Extraction. The accuracy of the classification depends on the Feature Extraction stage. The different features that can be extracted for a digital mammogram are: Texture Features, Statistical Features, Structural Features. In this paper, we are able to calculate Texture, Statistical and Structural Features. We have used MATLAB for extracting the tumors from input mammogram and for calculating various features.

Suitable features selection for monitoring thermal condition of electrical equipment using infrared thermography

“…Energy is the sum of square elements in the grey level co-occurrence matrix. Contrast is a measure of the intensity contrast between a pixel and its neighbor over the whole image and entropy is used to calculate statistical randomness [31] [35]. Tables 4 and 5 show the training and testing performance of GLCM features for MLP networks trained using LM, SCG, BR and RP algorithms respectively.…”

“…Entropy measures the randomness of the input image. Variance determines the deviation of grey level pixels from the mean within the image and standard deviation of pixel intensities is as like as variance, but determines mean square deviation of grey level value from its mean value [30,31]. Tables 2 and 3 show the training and testing performance of first order histogram features for MLP networks trained using LM, SCG, BR and RP algorithms respectively.…”

An Appraisal of Content-Based Image Retrieval (CBIR) Methods

“…Features are observable patterns in the image that contain relevant information of an image. Pradeep et al [11] also viewed a feature as a piece of information that is relevant for solving a computational task related to a certain application. Features describe and define the content of an image.…”

Automatic Feature Extraction for Breast Density Segmentation and Classification

“…Mammographic risk scoring refers to the representation of the mammograms to the risk of cancer [20]. There are various scales to represent the various levels of the spread of cancerous tissues in the body.…”