Related Experiment Video
Updated: Jun 19, 2026
10:25
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
Published on: September 25, 2019
Interrater agreement for final infarct MRI lesion delineation
Anders B Neumann1, Kristjana Y Jonsdottir, Kim Mouridsen
1Center of Functionally Integrative Neuroscience, Aarhus University Hospitals, Aarhus, Denmark. neumann@pet.auh.dk
Stroke
|October 3, 2009
Summary
Fluid-attenuated inverse recovery (FLAIR) MRI provides better interrater agreement for stroke lesion volume measurement than T2-weighted MRI. This finding is crucial for improving the reliability of clinical stroke trial outcomes.
Area of Science:
- Neurology
- Radiology
- Medical Imaging
Background:
- Magnetic resonance imaging (MRI) lesion volume is a key outcome in stroke trials.
- Optimal MRI sequence choice and interrater reliability for lesion volume assessment require further evaluation.
Purpose of the Study:
- To quantify interrater agreement for chronic infarct delineation using T2-weighted and fluid-attenuated inverse recovery (FLAIR) MRI sequences.
Main Methods:
- Retrospective analysis of 14 patients with chronic infarcts.
- Nine neuroradiologists delineated lesions on 90-day follow-up FLAIR and T2 fast spin echo MRI.
- Interrater agreement assessed using overlap ratio and Hausdorff distance.
Main Results:
- FLAIR sequences showed significantly better interrater agreement than T2 sequences.
- Overlap ratio was greater and Hausdorff distance was smaller on FLAIR compared to T2 (P<0.0001).
- Mean lesion volume was 2.5 mL greater on FLAIR than T2.
Conclusions:
- FLAIR MRI is recommended for quantifying follow-up infarct size to reduce interrater variability in stroke clinical trials.
- Consideration of single or few trained readers for imaging analysis is suggested.
- Future research should focus on objective and automated lesion delineation criteria.
