Biomedical Engineering Reference
In-Depth Information
A Sparse-modeled ROI for GLAM
Construction in Image Classication
Problems
—
A Case Study of Breast Cancer
K. Karteeka Pavan and Ch. Srinivasa Rao
Abstract Image segmentation is a process to determine regions of interest (ROI) in
mammograms. Mammograms can be classi
ed by extracting textural features of
ROI using Gray Level Aura Matrices (GLAM). Scientists are selecting a
xed
window size for all ROIs to
nd respective GLAM, though the masses will not
occur in regular two-dimensional geometries. This paper makes an attempt to
replicate the problem but by choosing arbitrary shape of masses as they occur. It is
found that this kind of natural selection of the arbitrary shape yielded drastic
reduction in time complexity by adopting the method suggested by us.
Keywords Classi
cation
Mammogram
Texture
GLAM
Segmentation
ROI
1 Introduction
Breast cancer places a predominant role in women mortality [
1
]. Hence, this subject
has become one of the hot topics in the study of image mining [
2
,
3
]. Hither to
scientists are choosing region of interest (ROI) as regular geometries while the fact
is that the mass region will always be of arbitrary shape. For this reason, in their
study, unconsciously, they are considering more number of unwanted pixels
increasing time complexity in extracting features using gray level matrices [
4
,
5
].
Therefore, in this paper, it is planned not to consider the unwanted pixels in feature
extraction. The notion of sparse matrix is conveniently made used in extracting
features based on the same technique of Gray Level Aura Matrices (GLAM) as
Search WWH ::
Custom Search