Event Date: Jul 25
2013 - 10:00am - 11:00am
Location: ICT 516, 856 Campus Place, University of Calgary.
Description: Breast masses exhibit a great variability in margins, shapes, and dimensions,
and their detection represents a difficult task to be solved for computer-aided
detection (CAD) of breast cancer with mammography. Masses detection is usually a
two-step procedure: mass identification and false-positive reduction. We propose
a method to automatically detect masses in mammographic images. The algorithm is
tuned according to the breast tissue density in order to improve the overall
performance in breast mass detection. The method uses modified phase portrait
analysis, based on the condition number (CN), a new eigenvalue intensity map
(EIM), a novel iterative and tissue-density-adaptive segmentation procedure, and
extraction of geometric features. Read more here.
No comments:
Post a Comment