Textural content in 3T MR: An image-based marker for Alzheimer's disease

Bharath Kumar Sriperumbudur, Rakesh Mullick, Uday Patil

Research output: Contribution to journalConference articlepeer-review

3 Scopus citations

Abstract

In this paper, we propose a study, which investigates the first-order and second-order distributions of T2 images from a magnetic resonance (MR) scan for an age-matched data set of 24 Alzheimer's disease and 17 normal patients. The study is motivated by the desire to analyze the brain iron uptake in the hippocampus of Alzheimer's patients, which is captured by low T 2 values. Since, excess iron deposition occurs locally in certain regions of the brain, we are motivated to investigate the spatial distribution of T2, which is captured by higher-order statistics. Based on the first-order and second-order distributions (involving gray level co-occurrence matrix) of T2 we show that the second-order statistics provide features with sensitivity > 90% (at 80% specificity), which in turn capture the textural content in T2 data. Hence, we argue that different texture characteristics of T2 in the hippocampus for Alzheimer's and normal patients could be used as an early indicator of Alzheimer's disease.

Original languageEnglish (US)
Article number156
Pages (from-to)1366-1376
Number of pages11
JournalProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume5747
Issue numberII
DOIs
StatePublished - Aug 25 2005

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Atomic and Molecular Physics, and Optics
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Fingerprint

Dive into the research topics of 'Textural content in 3T MR: An image-based marker for Alzheimer's disease'. Together they form a unique fingerprint.

Cite this