Related Experiment Videos
Problems with ratio and proportion measures of imaged cerebral structures
S Arndt1, G Cohen, R J Alliger
1Department of Preventive Medicine and Environmental Health, University of Iowa College of Medicine, Iowa City.
Psychiatry Research
|May 1, 1991
Summary
Ratio measures in brain imaging research, like the ventricle-brain ratio (VBR), can be unreliable. Simple volume measures or regression methods may offer more dependable brain structure assessments.
Area of Science:
- Neuroimaging
- Psychiatric Research
- Brain Morphology
Background:
- Ratio measures, such as the ventricle-brain ratio (VBR), are common in psychiatric research using CT and MRI.
- These ratios, comparing a structure's size to whole brain size, have persisted despite imaging advancements.
Purpose of the Study:
- To evaluate the reliability of ratio measures compared to simple volume measures in brain imaging.
- To discuss alternative methods for adjusting brain structure size for whole brain size.
Main Methods:
- Demonstrated the unreliability of ratio measures.
- Reviewed various methods for measuring brain structure size.
- Discussed strengths and limitations regarding reliability and validity.
Main Results:
- Ratio measures can suffer significant reliability issues compared to simple volume measures.
- This loss of reliability is linked to the relationship between a structure's size and overall brain size.
Conclusions:
- Ratio measures may be less reliable than simple volume measures for assessing brain structures.
- Methods like regression or covariance analysis may provide more appropriate brain size adjustments.