TY  - Conference Object
AU  - Galdran, Adrian
AU  - Picon, Artzai
AU  - Garrote, Estibaliz
AU  - Pardo, David
TI  - Pectoral Muscle Segmentation in Mammograms Based on Cartoon-Texture Decomposition
PY  - 2015
PB  - Springer
AB  - Pectoral muscle segmentation on medio-lateral oblique views
of mammograms represents an important preprocessing step in many
mammographic image analysis tasks. Although its location can be per-
ceptually obvious for a human observer, the variability in shape, size, and
intensities of the pectoral muscle boundary turns its automatic segmen-
tation into a challenging problem. In this work we propose to decompose
the input mammogram into its textural and structural components at
di erent scales prior to dynamically thresholding it into several levels.
The resulting segmentations are re ned with an active contour model and
merged together by means of a simple voting scheme to remove possible
outliers. Our method performs well compared to several other state-of-
the-art techniques. An average DICE similarity coe cient of 0:91 and
mean Hausdor  distance of 3:66   3:23 mm. validate our approach.
UR  - http://hdl.handle.net/11556/200
SN  - 978-3-319-19390-8
ER  -