AM-FM texture image analysis in multiple sclerosis brain white matter lesions

C. P. Loizou, V. Murray, M. S. Pattichis, M. Pantziaris, I. Seimenis, C. S. Pattichis

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

Abstract

In this study we investigate the use of multiscale Amplitude Modulation-Frequency Modulation (AM-FM) methods for analyzing brain white matter lesions that are associated with multiple sclerosis MRI lesions imaged at 0 and 6-12 months. We use the instantaneous amplitude (IA) and the instantaneous frequency (IF) to assess disease progression. The IA and the IF were calculated in transverse sections of T2- weighted magnetic resonance (MR) images acquired from 38 symptomatic untreated subjects between the first and the second examination scan. The findings suggest that the high-, medium-, and low- frequency scale instantaneous amplitude and frequency can be used to differentiate between normal tissue and lesions at 0 and 6-12 months. Moreover, support vector machine (SVM) models gave satisfactory results for differentiating lesions at 0 months using the medium scale IA and IF components for expanded disability status scale (EDSS) <=2 and EDSS >2. Further work is needed with more subjects in validating the proposed AM-FM analysis.

Original languageEnglish
Title of host publicationXII Mediterranean Conference on Medical and Biological Engineering and Computing 2010, MEDICON 2010
Pages446-449
Number of pages4
Volume29
DOIs
Publication statusPublished - 2010
Event12th Mediterranean Conference on Medical and Biological Engineering and Computing, MEDICON 2010 - Chalkidiki, Greece
Duration: 27 May 201030 May 2010

Other

Other12th Mediterranean Conference on Medical and Biological Engineering and Computing, MEDICON 2010
Country/TerritoryGreece
CityChalkidiki
Period27/05/1030/05/10

Keywords

  • AM-FM analysis
  • brain white matter
  • disease progression
  • Multiple sclerosis

Fingerprint

Dive into the research topics of 'AM-FM texture image analysis in multiple sclerosis brain white matter lesions'. Together they form a unique fingerprint.

Cite this