Related Experiment Video
Updated: Dec 16, 2025
Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images
Published on: January 7, 2019
Reduced accuracy of MRI deep grey matter segmentation in multiple sclerosis: an evaluation of four automated methods
Alexandra de Sitter1, Tom Verhoeven2, Jessica Burggraaff3
1Department of Radiology and Nuclear Medicine, MS Center Amsterdam, Amsterdam Neuroscience, Amsterdam UMC, Location VUmc, De Boelelaan 1117, 1081 HV, Amsterdam, The Netherlands. A.deSitter@amsterdamumc.nl.
Background:
Deep grey matter (DGM) atrophy in multiple sclerosis (MS) and its relation to cognitive and clinical decline requires accurate measurements. MS pathology may deteriorate the performance of automated segmentation methods. Accuracy of DGM segmentation methods is compared between MS and controls, and the relation of performance with lesions and atrophy is studied.
Methods:
On images of 21 MS subjects and 11 controls, three raters manually outlined caudate nucleus, putamen and thalamus; outlines were combined by majority voting. FSL-FIRST, FreeSurfer, Geodesic Information Flow and volBrain were evaluated. Performance was evaluated volumetrically (intra-class correlation coefficient (ICC)) and spatially (Dice similarity coefficient (DSC)). Spearman's correlations of DSC with global and local lesion volume, structure of interest volume (ROIV), and normalized brain volume (NBV) were assessed.
Results:
ICC with manual volumes was mostly good and spatial agreement was high. MS exhibited significantly lower DSC than controls for thalamus and putamen. For some combinations of structure and method, DSC correlated negatively with lesion volume or positively with NBV or ROIV. Lesion-filling did not substantially change segmentations.
Conclusions:
Automated methods have impaired performance in patients. Performance generally deteriorated with higher lesion volume and lower NBV and ROIV, suggesting that these may contribute to the impaired performance.
Insights
Automated deep grey matter segmentation methods perform less accurately in multiple sclerosis (MS) patients compared to controls. Performance declines with increased lesion volume and reduced brain volume, impacting cognitive decline assessments.
Area of Science:
- Neuroimaging
- Medical Image Analysis
- Multiple Sclerosis Research
Background:
- Accurate measurement of deep grey matter (DGM) atrophy is crucial for understanding cognitive and clinical decline in multiple sclerosis (MS).
- MS-related pathology can negatively impact the performance of automated segmentation techniques used for DGM analysis.
- This study investigates the accuracy of DGM segmentation methods in MS patients versus controls and examines the influence of lesions and atrophy on performance.
Purpose of the Study:
- To compare the accuracy of automated deep grey matter segmentation methods between multiple sclerosis patients and healthy controls.
- To assess the relationship between segmentation performance and the presence of lesions and brain atrophy in MS.
- To evaluate the impact of MS pathology on the reliability of automated DGM volume measurements.
Main Methods:
- Manual segmentation of caudate nucleus, putamen, and thalamus by three raters in 21 MS patients and 11 controls.
- Evaluation of four automated segmentation tools: FSL-FIRST, FreeSurfer, Geodesic Information Flow, and volBrain.
- Performance assessment using volumetric (ICC) and spatial (DSC) metrics, with correlation analysis against lesion volume, structure volume (ROIV), and normalized brain volume (NBV).
Main Results:
- Automated methods showed generally good volumetric agreement (ICC) but varied spatial agreement (DSC) with manual segmentations.
- Significantly lower DSC was observed for thalamus and putamen in MS patients compared to controls.
- Segmentation accuracy (DSC) negatively correlated with lesion volume and positively with NBV and ROIV in some cases.
Conclusions:
- Automated deep grey matter segmentation methods exhibit impaired performance in individuals with multiple sclerosis.
- Segmentation performance is generally reduced with increasing lesion burden and decreasing normalized brain volume and region of interest volume.
- Lesion-filling techniques did not significantly alter segmentation outcomes, indicating that MS-related changes directly impact automated method accuracy.
