Title :
Detection of spiculated lesions in mammograms
Author :
Sampat, Mehul P. ; Bovik, Alan C.
Author_Institution :
Dept. of Biomedical Eng., Texas Univ., Austin, TX, USA
Abstract :
In this paper, we present a new technique for the detection of spiculated masses in digitized mammograms. The techniques consist of two stages, enhancement of spiculations followed by the detection of the location where they converge. We describe a new algorithm for the enhancement and a new set of linear image filters which we have created for the detection stage. We have tested the algorithm on digitized mammograms obtained from the digital database for screening mammography (DDSM). Results of the detection algorithm are shown. Finally we show that the algorithm may be modified for the detection of architectural distortions.
Keywords :
Radon transforms; cancer; mammography; medical image processing; DDSM; digital database for screening mammography; image filters; mammograms; spiculated lesions detection; spiculation enhancement; Biomedical computing; Biomedical engineering; Breast cancer; Cancer detection; Filtering algorithms; Filtering theory; Image converters; Lesions; Mammography; Nonlinear filters;
Conference_Titel :
Engineering in Medicine and Biology Society, 2003. Proceedings of the 25th Annual International Conference of the IEEE
Print_ISBN :
0-7803-7789-3
DOI :
10.1109/IEMBS.2003.1279888