Related Experiment Videos
Determining the OPTIMAL DTI analysis method for application in cerebral small vessel disease.
Marco Egle1, Saima Hilal2, Anil M Tuladhar3
1Stroke Research Group, Department of Clinical Neurosciences, University of Cambridge, Cambridge, United Kingdom.
Neuroimage. Clinical
|July 31, 2022
Summary
Diffusion tensor imaging (DTI) analysis methods for cerebral small vessel disease (SVD) can predict dementia conversion. While several strategies show promise, automated methods like peak width skeletonized mean diffusivity (PSMD) are advantageous for large clinical trials.
Area of Science:
- Neuroimaging
- Neurology
- Biomedical Engineering
Background:
- Diffusion tensor imaging (DTI) shows potential as a surrogate marker for white matter (WM) damage in cerebral small vessel disease (SVD).
- Optimizing DTI analysis is crucial for its application in phase 2 clinical trials for SVD.
- The ideal DTI analysis method should be sensitive to change, predict dementia conversion, and be automated.
Purpose of the Study:
- To evaluate five different DTI analysis strategies for their utility in SVD clinical trials.
- To determine which DTI analysis method is most sensitive to change and best predicts dementia conversion in SVD cohorts.
- To assess the feasibility of implementing these DTI methods in clinical trials, including sample size estimations.
Main Methods:
- Five DTI analysis strategies were evaluated: mean diffusivity (MD) median, principal component 1 (PC1), peak width skeletonized mean diffusivity (PSMD), diffusion tensor image segmentation θ (DSEG θ), and global network efficiency (Geff).
- Associations between DTI measures and cognitive function were assessed using linear regression.
- Prediction of dementia conversion was examined using Cox proportional-hazard or logistic regression models in three cohort studies with longitudinal data.
- Sample size calculations for hypothetical clinical trials were performed for each strategy.
Main Results:
- All DTI measures showed a cross-sectional association with impaired cognitive function.
- Baseline DTI measures predicted dementia conversion in severe SVD, mild SVD, and MCI (except Geff in MCI).
- Automated measures PSMD and DSEG θ required the smallest sample sizes for hypothetical trials in sporadic SVD cohorts.
Conclusions:
- All evaluated DTI analysis methods can predict dementia in SVD.
- No single DTI analysis strategy emerged as definitively superior for all aspects of SVD clinical trials.
- Fully automated methods, such as PSMD, offer practical advantages for analyzing large datasets in SVD research.