Research Article

Computerized Segmentation and Characterization of Breast Lesions in Dynamic Contrast-Enhanced MR Images Using Fuzzy c-Means Clustering and Snake Algorithm

Table 1

Areas, statistical comparisons and area overlap measures of computerized delineation and radiologists’ manual delineation.

Segmentation methodArea
(mean ± SD pixels)
Pearson’s correlationt-test
P value
AOR1
(mean ± SD)
AOR2
(mean ± SD)

FCM-based1599.5 ± 1355.40.8910.1050.75 ± 0.130.72 ± 0.12
GVF-FCM1815.3 ± 1722.20.9760.4370.81 ± 0.100.78 ± 0.08
Radiologists’ manual2114.9 ± 2093.8