Name
Abstract | ||
|---|---|---|
Automatic segmentation of adipose tissue in thigh magnetic resonance imaging (MRI) scans is challenging and rarely reported in the literature. To address this problem, we propose a fully automated unsupervised segmentation method involving the use of spatial intensity constraints to guide the segmentation process. The novelty of this method lies in two aspects: firstly, an adaptive distance classifier, incorporating intra-slice spatial continuity, is used for robust region growing and segmentation estimation; secondly, polynomial based intensity inhomogeneity maps are generated to model inter-and intra-slice intensity variation of each pixel class and thus refine the initial classification. Our experimental results have demonstrated the effectiveness of imposing 3D intensity constraints to successfully classify the adipose tissue from muscles in the presence of image noise and considerable amounts of non-uniform MRI intensity. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1007/978-3-642-39094-4_51 | IMAGE ANALYSIS AND RECOGNITION |
Keywords | Field | DocType |
Adipose tissue, thigh MRI, intensity inhomogeneity | Computer vision,Scale-space segmentation,Polynomial,Pattern recognition,Segmentation,Computer science,Image noise,Artificial intelligence,Region growing,Pixel,Classifier (linguistics),Magnetic resonance imaging | Conference |
Volume | ISSN | Citations |
7950 | 0302-9743 | 1 |
PageRank | References | Authors |
0.36 | 5 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Senthil Purushwalkam | 1 | 42 | 5.11 |
| Baihua Li | 2 | 176 | 21.71 |
| Qinggang Meng | 3 | 273 | 23.54 |
| Jamie S. McPhee | 4 | 23 | 2.92 |