Abstract
Multiple sclerosis (MS) is a common autoimmune disorder,
whose diagnosis and study often relies on the extraction
of biomarkers from magnetic resonance imaging (MRI)
scans. Manual segmentation of MS lesions suffers from large
intra- and inter-rater differences, whereas automatic methods
promise reproducibility and enhanced consistency, especially
for tracking the disease progress over time. To test this claim,
the ISBI 2015 Longitudinal MS Lesion Segmentation Challenge
provides a platform to compare existing methods in a
fair and consistent manner to each other and the manual approach.
In this article, we present our challenge contribution,
which is based on random forests and local context intensity
features to segment MS lesions in multi-spectral MRI images.
whose diagnosis and study often relies on the extraction
of biomarkers from magnetic resonance imaging (MRI)
scans. Manual segmentation of MS lesions suffers from large
intra- and inter-rater differences, whereas automatic methods
promise reproducibility and enhanced consistency, especially
for tracking the disease progress over time. To test this claim,
the ISBI 2015 Longitudinal MS Lesion Segmentation Challenge
provides a platform to compare existing methods in a
fair and consistent manner to each other and the manual approach.
In this article, we present our challenge contribution,
which is based on random forests and local context intensity
features to segment MS lesions in multi-spectral MRI images.
| Original language | English |
|---|---|
| Journal | Proc. 2015 Longitudinal Multiple Sclerosis Lesion Segmentation Challenge |
| Pages (from-to) | 1-2 |
| Number of pages | 2 |
| DOIs | |
| Publication status | Published - 2015 |
