Abstract
The objective of this paper is to investigate a potential approach of mammographic image segmentation based on textons and mammographic building blocks (i.e. nodular, linear, homogeneous, and radiolucent) as described in Tabar’ ´ s tissue model. The texton approach is based on clustering filter responses in a high dimension for a
particular building block. The texton selection process is based on a combination of visual assessment (probability maps) and minimum spanning tree topological information. The initial segmentation results are promising, which may lead us to an automatic Tabar’ ´ s mammographic risk assessment
particular building block. The texton selection process is based on a combination of visual assessment (probability maps) and minimum spanning tree topological information. The initial segmentation results are promising, which may lead us to an automatic Tabar’ ´ s mammographic risk assessment
| Original language | English |
|---|---|
| Pages | 228-232 |
| Number of pages | 5 |
| Publication status | Published - 17 Jul 2007 |
| Event | Medical Image Understanding and Analysis 2007: University of Wales Aberystwyth, 17-18 th July - University of Wales, Aberystwyth, Aberystwyth, United Kingdom of Great Britain and Northern Ireland Duration: 17 Jul 2007 → 18 Jul 2007 |
Conference
| Conference | Medical Image Understanding and Analysis 2007 |
|---|---|
| Abbreviated title | MIUA 2007 |
| Country/Territory | United Kingdom of Great Britain and Northern Ireland |
| City | Aberystwyth |
| Period | 17 Jul 2007 → 18 Jul 2007 |