Title :
Shape analysis of mammographic calcifications
Author :
Shen, L. ; Rangayyan, Rangaraj M. ; Desautels, J. E Leo
Author_Institution :
Dept. of Electr. & Comput. Eng., Calgary Univ., Alta., Canada
Abstract :
A set of shape factors has been developed to measure the roughness of regions of calcifications in mammograms, to aid in their classification as malignant or benign. The analysis of mammograms is performed in three stages. First, a region growing technique is used to obtain the contours of regions including calcifications. Then, measures of shape features, such as compactness, moments, and Fourier descriptors, are computed for each region. Finally, their applicability to classification is studied by using the above shape measures to form feature vectors
Keywords :
Fourier analysis; diagnostic radiography; medical diagnostic computing; medical image processing; Fourier descriptors; breast cancer; form feature vectors; mammographic calcifications; medical diagnostic imaging; region growing technique; shape analysis; shape factors; Breast cancer; Cancer detection; Diseases; Electric variables measurement; Hospitals; Image processing; Mammography; Performance analysis; Radiology; Shape measurement;
Conference_Titel :
Computer-Based Medical Systems, 1992. Proceedings., Fifth Annual IEEE Symposium on
Conference_Location :
Durham, NC
Print_ISBN :
0-8186-2742-5
DOI :
10.1109/CBMS.1992.244955