Content based retrieval of lesioned brain images.
Presented at: UNSPECIFIED.
HI-PACS enable more efficient data-management leading to increased operating efficiency and therefore better patient care, a content based pet image retrieval system would contribute to the development of a HI-PACS. A database of PET neuro-images has been created with a facility for retrieving via visual content. The adaptation of algorithms developed for alternate imaging modalities (eg-MRI) formed the basis of feature detection and measurement algorithms. The application of these algorithms to greyscale PET images results in data that is employed as database indices and similarity metrics. The feature detection and measurement algorithms can be split into two different methods. The first uses the extracted ideal mid sagittal symmetry line to detect differences between the two hemisphere of the brain, while the second utilizes Gabor filters to measure the texture of the whole brain.
|Type:||Conference item (UNSPECIFIED)|
|Title:||Content based retrieval of lesioned brain images|
|UCL classification:||UCL > School of BEAMS > Faculty of Engineering Science
UCL > School of BEAMS > Faculty of Engineering Science > Computer Science
Archive Staff Only