Automated MR image processing and analysis of malignant brain tumors: Enabling technology for data mining

Shishir Dube, Jason J. Corso, Timothy F. Cloughesy, Suzie El-Saden, Alan L. Yuille, Usha Sinha

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Scopus citations

Abstract

Glioblastoma multiforme (GBM) is a malignant brain cancer with poor patient prognosis (i.e. time to survival, time to tumor progression). A number of clinical trials are underway evaluating novel therapeutic strategies and magnetic resonance imaging is the most routinely performed procedure for accurate serial monitoring of patients. The electronic availability of the comprehensive data collected as part of the clinical trials provides an unprecedented opportunity to discover new relationships in complex diseases such as GBM. However, imaging data, which is the most accurate non-invasive assessment of GBMs, is not directly amenable for data mining. The focus of this chapter is on image analysis techniques including image spatial and intensity standardization, novel methods for robust tumor and edema segmentation, and quantification of tumor intensity, texture, and shape characteristics. The chapter concludes with an application of discovering the relationship between these quantitative image-derived features and time to survival in GBM patients; the data is part of a comprehensive large electronically accessible archive at UCLA (UCLA Neuro-oncology database).

Original languageEnglish
Title of host publicationData Mining, Systems Analysis, and Optimization in Biomedicine
Pages64-84
Number of pages21
DOIs
StatePublished - 2007
EventConference on Data Mining, Systems Analysis, and Optimization in Biomedicine, 2007 - Gainesville, FL, United States
Duration: 28 Mar 200730 Mar 2007

Publication series

NameAIP Conference Proceedings
Volume953
ISSN (Print)0094-243X
ISSN (Electronic)1551-7616

Conference

ConferenceConference on Data Mining, Systems Analysis, and Optimization in Biomedicine, 2007
Country/TerritoryUnited States
CityGainesville, FL
Period28/03/0730/03/07

Keywords

  • Classification
  • Data mining
  • Image segmentation
  • Pattern recognition

Fingerprint

Dive into the research topics of 'Automated MR image processing and analysis of malignant brain tumors: Enabling technology for data mining'. Together they form a unique fingerprint.

Cite this