Mammograms taken at prescribed intervals fail to curtail the vast occurrence of breast cancers. One foremost cause is that at the initial stages where there are only minor symptoms visible in a mammogram chances are that it could be overlooked during perusal. There are many signs of breast cancer like Calcification, Masses, Bilateral Asymmetry and Architectural Distortion. Architectural Distortion may be associated with early diagnosis of breast cancer because even before there is a visible mass, cancer growth can disrupt parenchyma structure. Double reading of screening mammograms could provide higher sensitivity over single reading, but the limitation on time and trained professionals makes it a not so possible approach. Algorithms are developed to assist radiologists in detecting abnormalities in mammograms. In this paper, a system is developed to classify Architectural Distortion abnormality from normal mammogram samples. Gabor features along with Law’s Texture Energy measures derived from geometrically transformed regions of interests are used to detect architectural distortion. The method has a good potential in detecting architectural distortion in mammograms of interval cancer cases.