3.2 Results
Experimental data is 60 sets abdominal CT image
obtained from a large hospital in Shenyang of China,
among them, 40 sets are served as training sample and
20 sets are acted as test sample. The training sample
consists of 20 sets of normal data and 20 sets of
abnormal data, while the test sample is composed of 10
sets of normal data and 10 sets of abnormal data.
A pancreas segmentation based on the level set
method guided by the SSM conducted to improve the
segmentation accuracy. The segmentation result is
shown in Fig. 2.