Skip to main content

Table 2 Description and name of PET-AS methods used in this study. The references correspond to recent publications using similar PET-AS algorithms

From: A novel phantom technique for evaluating the performance of PET auto-segmentation methods in delineating heterogeneous and irregular lesions

Segmentation approach Name Description
Adaptive thresholding AT 3D iterative background-subtracted thresholding
Region-growing RG 3D iterative region-growing with automatic seed finder
Clustering GCM2-GCM8 3D fuzzy C-means segmentation with Gaussian mixture modelling, identifying 2, 3, 4, 5, 6, 7 or 8 clusters
Watershed Transform WT Slice-by-slice watershed transform-based segmentation, with automatic seed finder