Blair Mackiewich. Intracranial boundary detection and radio frequency correction in magnetic resonance images. Master's thesis, Simon Fraser University, Computer Science Department, Burnaby, B.C., August 1995.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....a semi automatic technique. As a part of the pre processing of the image data for segmentation, identification of the contour of the brain surface inside the intracranial cavity is required. Recent research has provided an automatic method for detecting this contour in an arbitrary MRI volume [45]. Since the data outside the brain contour is of no use to the segmentation, this provides a means of partitioning the volume into disjoint subimages of differing relative importance, and this information can be used to improve compression. Most image compression systems are general purpose and ....

....be corrected using digital filtering techniques [20] It is useful, both for the correction and the segmentation processes, to isolate the brain in each slice from the surrounding tissues such as skull, muscle and fat. Conventionally, this is performed manually, as in [6] but recently Mackiewich [45] has obtained very good results using automatic techniques. Furthermore, postprocessing of the partial volume segmentation can be used to provide an improved segmentation of lesions alone. The final result of the segmentation is a binary classification of each voxel in the volume as lesion or ....

[Article contains additional citation context not shown here]

Blair T. Mackiewich. Intracranial boundary detection and radio frequency correction in magnetic resonance images. M.Sc. thesis, Simon Fraser University, August 1995.

....relatively high intensity and when the T2 (or PD) image is filtered with a special anisotropic diffusion filter, the non brain tissues can be attenuated, and hence the brain tissues can be segmented using a simple threshold. The methodology is described in more detail below, and in full detail in [20]. Although seemingly complex, the method has proven so robust that it works even in the presence of RF inhomogeneity, which would not be the case if a simple threshold technique were used to identify brain tissue from non brain tissue. Similar techniques have been applied to chest MRI to segment ....

....distance from the pixel of interest [10] Anisotropic data can be handled similarly. 2 We also attempted to attenuate non brain tissues using 3D nonlinear anisotropic diffusion over the entire MRI volume. 3D diffusion, however, increased the partial volume effect, blurring the edges of the brain [20]. 2 METHOD 7 (a) b) Figure 4: Intracranial boundary detection using 2D nonlinear anisotropic diffusion filtering. a) Original image. b) 2D diffused image. Diffusion attenuates non brain tissues, enabling a simple threshold to segment the brain. 2 METHOD 8 best fit Gaussian curve histogram ....

[Article contains additional citation context not shown here]

Blair Mackiewich. Intracranial boundary detection and radio frequency correction in magnetic resonance images. Master's thesis, Simon Fraser University, Computer Science Department, Burnaby, B.C., August 1995.

No context found.

Blair T. Mackiewich. Intracranial boundary detection and radio frequency correction in magnetic resonance images. M.Sc. thesis, Simon Fraser University, August 1995.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC