|
KMID : 1147120140200010019
|
|
Journal of the Korean Society of Imaging Informatics in Medicine
2014 Volume.20 No. 1 p.19 ~ p.26
|
|
|
Opportunities and Challenges in Radiogenomics: Imaging Phenotype Analysis for Brain Tumor
|
|
Lee Myung-Eun
Kim Jong-Hyo
|
|
|
Abstract
|
|
|
|
Radiogenomics is an emerging field of study which aims at improving cancer treatment outcome by discovering and employing associations between genomic and imaging phenotype signatures in various cancers. The TCGA project launched in 2006 enabled active researches in radiogenomics, which has been hindered by the lack of accessibility to high quality and high volume genomic and imaging data obtained at same patients. The TCGA project was initiated in order to promote the basic understanding the molecular biology of cancer in accordance of rapidly developing genome sequencing technologies and therewith amassing genomic data of tumors, which allowed collection of large amount of genomic and imaging data through a standardized protocol from a number of institutions. In particular, the imaging data having the same unique identifier as TCGA data are archived at TCIA site, and are accessible freely for researchers investigating the imaging phenotypes of cancer. High throughput analysis of imaging phenotype requires a set of appropriate and advanced image processing techniques including image registration, tumor segmentation, and feature extraction. Patient motion occurred during images acquisition requires image registration step, followed by tumor segmentation step and subsequent feature extraction step. Finally, associations between imaging phenotype and genomic data are analyzed through an advanced cluster and statistical testing. As an emerging field, landscape of radiogenomic continues to change and many challenging issues remain unexplored. When the imaging phenotype and genotype signatures of cancers are appropriately integrated into a single platform in the future, the radiogenomics platform could enable the subtype identification of given cancer along with survival prediction for a set of alternative treatment options, which is the promise of personalized medicine.
|
|
|
KEYWORD
|
|
|
Radiogenomics, Imaging Phenotype, The Cancer Genome Atlas (TCGA), The Cancer Imaging Archive (TCIA), Segmentation, Glioblastoma Multiforme (GBM), Feature Extraction
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|
|