High intensity region segmentation in MR imaging of multiple sclerosis

Numerous pathologies are often manifest in Magnetic Resonance Imaging (MRI) as hyperintense or bright regions as compared to normal tissue. It is of particular interest to develop an algorithm to detect, identify and define those Regions of Interest (ROI) when analyzing MRI studies, particularly for...

Full description

Saved in:
Bibliographic Details
Published in:Journal of physics. Conference series 2013-01, Vol.477 (1), p.12024-7
Main Authors: Rodrigo, F, Filipuzzi, M, Isoardi, R, Noceti, M, Graffigna, J P
Format: Article
Language:English
Subjects:
Algorithms
Bones
Data mining
Image segmentation
Lesions
Magnetic resonance imaging
Medical imaging
Multiple sclerosis
Physics
Segmentation
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Numerous pathologies are often manifest in Magnetic Resonance Imaging (MRI) as hyperintense or bright regions as compared to normal tissue. It is of particular interest to develop an algorithm to detect, identify and define those Regions of Interest (ROI) when analyzing MRI studies, particularly for lesions of Multiple Sclerosis (MS). The objective of this study is to analyze those parameters which optimize segmentation of the areas of interest. To establish which areas should be considered as hyperintense regions, we developed a database (DB), with studies of patients diagnosed with MS. This disease causes axonal demyelination and it is expressed as bright regions in PD, T2 and FLAIR MRI sequences. Thus, with more than 4300 hyperintense regions validated by an expert physician, an algorithm was developed to detect such spots, approximating the results the expert obtained. Alongside these hyperintense lesion regions, it also detected bone regions with high intensity levels, similar to the intensity of the lesions, but with other features that allow a good differentiation.The algorithm will then detect ROIs with similar intensity levels and performs classification through data mining techniques.
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/477/1/012024