You are here
The Identification and Segmentation of Astrocytoma Prior to Critical Mass, by means of a Volumetric/Subregion Regression Analysis of Normal and Neoplastic Brain Tissue
| Title: | The Identification and Segmentation of Astrocytoma Prior to Critical Mass, by means of a Volumetric/Subregion Regression Analysis of Normal and Neoplastic Brain Tissue. |
 53 views 22 downloads |
|---|---|---|
| Name(s): |
Higgins, Lyn, Author Hughes, Charles, Committee Chair Morrow, Patricia Bockelman, Committee Member Bagci, Ulas, Committee Member Lisle, Curtis, Committee Member University of Central Florida, Degree Grantor |
|
| Type of Resource: | text | |
| Date Issued: | 2018 | |
| Publisher: | University of Central Florida | |
| Language(s): | English | |
| Abstract/Description: | As the underlying cause of Glioblastoma Multiforme (GBM) is presently unclear, this research implements a new approach to identifying and segmenting plausible instances of GBM prior to critical mass. Grade-IV Astrocytoma, or GBM, is an aggressive and malignant cancer arising from star-shaped glial cells, or astrocytes, where the astrocytes, functionally, assist in the support and protection of neurons within the central nervous system and spinal cord. Subsequently, our motivation for researching the ability to recognize GBM is that the underlying cause of the mutation is presently unclear, leading to the operative that GBM is only detectable through a combination of MRI and CT brain scans, cooperatively, along with a resection biopsy. Since astrocytoma only becomes evident at critical mass, when the cellular structure of the neoplasm becomes visible within the image, this research seeks to achieve earlier identification and segmentation of the neoplasm by evaluating the malignant area via a volumetric voxel approach to removing noise artifacts and analyzing voxel differentials. In order to investigate neoplasm continuity, a differential approach has been implemented utilizing a multi-polynomial/multi-domain regression algorithm, thus, ultimately, providing a graphical and mathematical analysis of the differentials within critical mass and non-critical mass images. Given these augmentations to MRI and CT image rectifications, we theorize that our approach will improve on astrocytoma recognition and segmentation, along with achieving greater accuracy in diagnostic evaluations of the malignant area. | |
| Identifier: | CFE0007336 (IID), ucf:52111 (fedora) | |
| Note(s): |
2018-12-01 M.S. Engineering and Computer Science, Dean's Office GRDST Masters This record was generated from author submitted information. |
|
| Subject(s): | Lyn Higgins -- GBM -- astrocytoma -- Lyn -- Higgins -- segmentation -- volumetric noise reduction -- 3D noise reduction -- multivariate Gaussian distribution -- noise reduction -- volumetric filtering -- volumetric analysis -- volumetric regression -- VVE -- VVK -- PVC -- volumetric voxel kernel -- volumetric voxel element -- primary voxel of concern -- R2n -- R3n -- critical mass -- pre-critical mass -- multi-temporal imaging -- volume correlations -- volume filtering -- multivariate Gaussian derivation -- volumetric architecture -- denoising -- noise reduction -- critical mass detection -- voxel regression -- voxel analysis -- neoplasm simulation -- GBM simulation -- astrocytoma simulation -- simulation -- voxel regression -- multi-polynomail -- multi-domain | |
| Persistent Link to This Record: | http://purl.flvc.org/ucf/fd/CFE0007336 | |
| Restrictions on Access: | campus 2019-12-15 | |
| Host Institution: | UCF |
