Using MRI from 1000 subjects to identify abnormal grey matter in individual tumor subjects

Atreya Misra, Rui Yuan, Suril Gohel, Bharat Biswal

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Detecting brain tumors is always an important but challenging task for MRI physicists and radiologists. Many automated brain tumor detection methods have been introduced. Due to the complexity and limitations of those procedures, practical applications of those methods in clinical setting are restricted. Based on a large sample size of acquired normal human datasets, this study provided a new approach to detect grey matter abnormalities in individual subjects. This approach consisted of three steps. First, voxel-based morphometry was performed on T1 images from 1000 normal subjects and each of the five tumor subjects. Second, we computed the distribution of the grey matter intensities of each voxel from 1000 normal subjects. Third, we compared the intensity of each grey matter voxel of tumor patient to its corresponding voxel's distribution of normal subjects. Finally, for each tumor subject, grey matter voxels that were out of 99.9% normal distribution were marked. This method uses large data sets to detect abnormalities and has the potential to be utilized in clinical applications.

Original languageEnglish (US)
Title of host publicationProceedings - 2014 40th Annual Northeast Bioengineering Conference, NEBEC 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781479937288
DOIs
StatePublished - Dec 2 2014
Event2014 40th Annual Northeast Bioengineering Conference, NEBEC 2014 - Boston, United States
Duration: Apr 25 2014Apr 27 2014

Publication series

NameProceedings of the IEEE Annual Northeast Bioengineering Conference, NEBEC
Volume2014-December
ISSN (Print)1071-121X
ISSN (Electronic)2160-7001

Other

Other2014 40th Annual Northeast Bioengineering Conference, NEBEC 2014
CountryUnited States
CityBoston
Period4/25/144/27/14

All Science Journal Classification (ASJC) codes

  • Bioengineering

Keywords

  • MRI
  • Tumor
  • VBM

Fingerprint Dive into the research topics of 'Using MRI from 1000 subjects to identify abnormal grey matter in individual tumor subjects'. Together they form a unique fingerprint.

Cite this