Skip to Main Content
 

Global Search Box

 
 
 
 

ETD Abstract Container

Abstract Header

Dynamic Co-occurrence of Local Anisotropic Gradient Orientations (DyCoLIAGe) Descriptors from Pre-treatment Perfusion DSC-MRI to Predict Overall Survival in Glioblastoma

Abstract Details

2019, Master of Sciences, Case Western Reserve University, Biomedical Engineering.
A significant clinical challenge in glioblastoma is to risk-stratify patients for clinical trials, preferably using MRI scans. Radiomics involves mining of sub-visual features that could serve as surrogate markers of tumor heterogeneity from routine imaging. Previously our group had developed a new gradient-based radiomic descriptor, Co-occurrence of Local Anisotropic Gradient Orientations (COLLAGE), to capture tumor heterogeneity on structural MRI. I present an extension of CoLLAGE on perfusion MRI, termed dynamic COLLAGE (DyCoLIAGe), and demonstrate its application in predicting overall survival in glioblastoma. Following manual segmentation, 52 CoLIAGe features were extracted from edema and enhancing tumor at different time phases during contrast administration of perfusion MRI. Each feature was separately plotted across different time-points, and a 3rd-order polynomial was fit to each feature curve. The corresponding polynomial coefficients were evaluated in terms of their prognosis performance. My results suggest that DyCoLIAGe may be prognostic of overall survival in glioblastoma.
Pallavi Tiwari (Committee Chair)
31 p.

Recommended Citations

Citations

  • Song, B. (2019). Dynamic Co-occurrence of Local Anisotropic Gradient Orientations (DyCoLIAGe) Descriptors from Pre-treatment Perfusion DSC-MRI to Predict Overall Survival in Glioblastoma [Master's thesis, Case Western Reserve University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=case1554299860994274

    APA Style (7th edition)

  • Song, Bolin. Dynamic Co-occurrence of Local Anisotropic Gradient Orientations (DyCoLIAGe) Descriptors from Pre-treatment Perfusion DSC-MRI to Predict Overall Survival in Glioblastoma. 2019. Case Western Reserve University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=case1554299860994274.

    MLA Style (8th edition)

  • Song, Bolin. "Dynamic Co-occurrence of Local Anisotropic Gradient Orientations (DyCoLIAGe) Descriptors from Pre-treatment Perfusion DSC-MRI to Predict Overall Survival in Glioblastoma." Master's thesis, Case Western Reserve University, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=case1554299860994274

    Chicago Manual of Style (17th edition)