COMPUTER AIDED SYSTEM FOR DETECTING MASSES IN MAMMOGRAMS

An intelligent CAD can be very helpful in detecting
masses in the breast earlier and faster than typical
screening programs. Two such systems are presented,
First a system based on Radial Basis neural networks
coupled with feature extraction techniques for
detecting masses in mammograms. Suspicious regions
are identified following a run of the trained neural
network. Co-occurrence matrices are constructed at
different distances for each mammogram. Statistical
features are used to train and test the Radial Basis
neural network. The second system presented was
developed based on linear subtraction and feature
extraction techniques to identify asymmetries
between left and right breast mammograms. This
system is based on the idea that a deviation from
the normal architectural symmetry of the right and
left breasts could indicate a cancerous mass. The
results show that both systems could be helpful to
the radiologist by serving as a second reader in
mammography screening.

Autorentext
Mohammed Jirari has received a B.S. in Computer Science with a minor in Mathematics from Edinboro University of Pennsylvania, a M.S. in Computer Science from Lamar University and a Ph.D. in Computer Science from Kent State University.

Klappentext
An intelligent CAD can be very helpful in detecting masses in the breast earlier and faster than typical screening programs. Two such systems are presented, First a system based on Radial Basis neural networks coupled with feature extraction techniques for detecting masses in mammograms. Suspicious regions are identified following a run of the trained neural network. Co-occurrence matrices are constructed at different distances for each mammogram. Statistical features are used to train and test the Radial Basis neural network. The second system presented was developed based on linear subtraction and feature extraction techniques to identify asymmetries between left and right breast mammograms. This system is based on the idea that a deviation from the normal architectural symmetry of the right and left breasts could indicate a cancerous mass. The results show that both systems could be helpful to the radiologist by serving as a second reader in mammography screening.