Yoo, Done-Sik; (1998) Imaging and segmentation of bone in neurological magnetic resonance. Doctoral thesis (Ph.D), UCL (University College London).

Text Imaging_and_segmentation_of_bo.pdf Download (13MB)

Abstract

Magnetic resonance imaging (MRI) is a modality capable of providing excellent contrast for soft tissues, yet it provides very little information about bone, which, by contrast, can be seen clearly in CT images. This project attempts to remedy this limitation by expanding the visualisation capability of MRI to include details of skull boundary. The potential benefits of this advance include the accurate co-registration of MRI and CT or MRI and MRI image data, as used in frameless surgery planning; the avoidance of harmful radiation, a problem encountered in CT skull visualisation; modelling of electrical conductivity in the head; and cranioplasty planning. In addition, the method developed will potentially be helpful for the measurement of brain volume, especially small changes in voxel size, which is used in the quantitative assessment of changes in neurological diseases. Skull edge detection is made difficult due to the partial volume effect, strong edges between muscle and scalp, the thin appearance of the skull in the temporal region and strong edges near the superior sagittal sinus. Imaging artefacts including chemical shift and magnetic susceptibility shift have also been investigated quantitatively and qualitatively. To detect skull contours, a new fully automatic computer algorithm has been developed and the results of the method have been validated.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item