Region Growing

Region growing [26] is a segmentation method used to extract a connected region of acceptably similar voxels from a scene. The similarity criteria are, in general, determined by a range of intensity values or by well-defined edges in the image. A seeded region growing requires an initialization seed, usually provided by the operator, to be placed within the target ROI. The algorithm then examines neighboring voxels, one at a time, adding those that fall within the acceptance criteria. Since each voxel is examined individually, the complete scene classification may be slow. In addition, region growing can also be sensitive to noise, and the segmentation of multiple regions requires at least one seed for each region to be classified. Because of the requirement of a manually selected seed, region growing is not well adapted for segmenting large, complex images such as MR acquisitions of the brain. However, it can be

FIGURE 1 All images depict the same brain, (a) Spin density and (b) T2-weighted double-echo images. Image parameters: TEs/TR = 34/100/3000ms, in-plane resolution = 0.94mm2, slice thickness = 5mm. (c) Tl-weighted spoiled grass (SPGR) image, TE/TR = 5/35 ms, in-plane resolution = 0.94 mm2, slice thickness = 1.5 mm. (d) Image shown in (c) segmented into white matter, gray matter, and cerebrospinal fluid with the adaptive Bayesian method. TE/TR = echo time/repetition time.

Aree Cerebrali Gliotiche

FIGURE 1 All images depict the same brain, (a) Spin density and (b) T2-weighted double-echo images. Image parameters: TEs/TR = 34/100/3000ms, in-plane resolution = 0.94mm2, slice thickness = 5mm. (c) Tl-weighted spoiled grass (SPGR) image, TE/TR = 5/35 ms, in-plane resolution = 0.94 mm2, slice thickness = 1.5 mm. (d) Image shown in (c) segmented into white matter, gray matter, and cerebrospinal fluid with the adaptive Bayesian method. TE/TR = echo time/repetition time.

useful when used in conjunction with thresholding for the classification of isolated tumors or lesions.

0 0

Post a comment