Abstract #1145
Minkowski Functionals in MRI: A new texture analysis tool in breast MRI
Michael Fox 1 , Peter Gibbs 1 , Martin Pickles 1 , and Lindsay W. Turnbull 1
1
Centre for MR Investigations, HYMS at
University of Hull, Hull, East Yorkshire, United Kingdom
Minkowski Functinals may provide a new method to observe
or diagnose breast cancer tumours in MRI, as they have
done in CT. Software was created in-house, using 101
threshold levels, to create binary images of segmented
breast lesions, from T1W MR images. Three 5th order
polynomial fits were created for each of the 100
patients to describe their change in MF values as the
threshold was raised. Two-thirds of the measured
parameters were able to distinguish between TNEG patient
sub-groups, with differences between biopsy grades being
found also. This work supports the introduction of
Minkowski Functionals into MRI texture analysis.
This abstract and the presentation materials are available to members only;
a login is required.
Join Here
