Related Experiment Video
Updated: Jul 21, 2026
Quantitative Magnetic Resonance Imaging of Skeletal Muscle Disease
Published on: December 18, 2016
Considerations for Mean Upper Cervical Cord Area Implementation in a Longitudinal MRI Setting: Methods, Interrater
C Chien1,2, V Juenger1,2,3, M Scheel3
1From the Experimental and Clinical Research Center (C.C., V.J., A.U.B., F.P.), Max Delbrück Center for Molecular Medicine & Charité-Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Berlin, Germany.
Technical variability in spinal cord atrophy measurements from MRI can be reduced using specific techniques. Non-distortion-corrected MRI and a semiautomatic segmentation approach improve reliability for upper cervical cord area analysis.
Area of Science:
- Neuroimaging and Quantitative MRI
- Spinal Cord Imaging and Analysis
- Medical Image Segmentation
Background:
- Spinal cord atrophy, particularly in the upper cervical cord, is a key metric assessed using cerebral MRI.
- Previous investigations have not thoroughly examined intraparticipant rescan variability in healthy cohorts for these measurements.
- Understanding technical and rescan variability is crucial for accurate longitudinal studies of spinal cord atrophy.
Purpose of the Study:
- To investigate the technical and rescan variability of mean upper cervical cord area measurements derived from T1-weighted cerebral MRIs.
- To compare variability between distortion-corrected and non-distortion-corrected MPRAGE sequences.
- To evaluate the impact of segmentation methods and rater variability on measurement accuracy.
Main Methods:
- Retrospective analysis of T1-weighted MPRAGE MRI scans from healthy participants.
- Inclusion of data from multiple scanning sessions (rescan variability) and daily variability studies.
- Utilized statistical methods including coefficient of variation, ICC, Bland-Altman analysis, and paired t-tests.
Main Results:
- Non-distortion-corrected MPRAGE sequences demonstrated robustness in real-world data, showing no significant difference in measures over time.
- Automatic segmentation methods resulted in higher variability compared to semiautomatic approaches.
- Interrater analysis revealed incomparable measures, highlighting the importance of a single rater and specific techniques.
Conclusions:
- Technical variability in spinal cord measurements can be minimized by employing non-distortion-corrected MRIs, a semiautomatic segmentation method, and a single rater.
- Rescan variability for group mean upper cervical cord area was found to be within ±4.4% when MRI quality criteria were met.
- Artifacts from patient movement, positioning, and metal implants are identified as primary sources of variability.
