Related Experiment Videos
Reliability of brain structure morphometry in hydrocephalic children using MR images
M E Brandt1, T P Bohan, K Thorstad
1Department of Psychiatry, University of Texas Medical School, Houston, USA 77030-1501. mbrandt@ped1.med.uth.tmc.edu
Magnetic Resonance Imaging
|January 1, 1996
Summary
This study shows that intraclass correlation coefficients (ICCs) reliably measure brain structures in children with hydrocephalus using MRI scans. These findings confirm that morphometric analysis is effective even in malformed brains.
Area of Science:
- Neuroimaging
- Pediatric Neurology
- Medical Statistics
Background:
- Assessing brain structure reliability in pediatric hydrocephalus is challenging.
- Morphometric measurements from MRI are crucial for understanding brain development and pathology.
- Operator variability can impact measurement accuracy in malformed brains.
Purpose of the Study:
- To evaluate the reliability of morphometric measurements of specific brain structures from MRI in children with hydrocephalus.
- To compare the effectiveness of Pearson's correlation coefficient (r) versus intraclass correlation coefficients (ICCs) for inter-rater reliability assessment.
- To determine if malformed brains, such as those in hydrocephalus cases, can be reliably assessed using MRI morphometrics.
Main Methods:
- Cross-sectional area and volume measurements of the corpus callosum, internal capsules, centrum semiovale, and lateral ventricles were performed.
- Measurements were taken from T1 and T2-weighted MRI scans of 50 children (hydrocephalus and controls) by two independent raters.
- Pearson's correlation coefficients (r) and intraclass correlation coefficients (ICCs) were computed to assess inter-rater reliability.
Main Results:
- Intraclass correlation coefficients (ICCs) ranged from 0.7502 to 0.9895, all significant at p < .0001.
- Pearson's r values were generally higher than ICCs, suggesting potential overestimation of reliability.
- The study confirmed that morphometric measures of MR images are reliable for assessing brains affected by disorders like hydrocephalus.
Conclusions:
- Intraclass correlation coefficients (ICCs) are recommended over Pearson's r for assessing inter-rater reliability, especially with significant group differences.
- Morphometric analysis of MRI is a reliable method for evaluating brain structures in children with hydrocephalus.
- Human operators can reliably determine region boundaries in significantly malformed brains using established morphometric techniques.