Related Experiment Video
Updated: Nov 2, 2025
11:50
A Standardized Pipeline for Examining Human Cerebellar Grey Matter Morphometry using Structural Magnetic Resonance Imaging
Published on: February 4, 2022
4.3K
Selecting software pipelines for change in flortaucipir SUVR: Balancing repeatability and group separation.
Christopher G Schwarz1, Terry M Therneau2, Stephen D Weigand2
1Department of Radiology, Mayo Clinic and Foundation, 200 First Street SW, Rochester 55905, MN, USA.
Neuroimage
|June 12, 2021
Summary
Optimizing tau PET quantification is key for Alzheimer's research. Longitudinal methods using larger temporal regions and specific reference areas improve accuracy and reduce sample size needs for clinical trials.
Area of Science:
- Neuroimaging
- Nuclear Medicine
- Alzheimer's Disease Research
Background:
- Quantification of tau Positron Emission Tomography (PET) tracers has utilized diverse automated methods.
- Serial tau PET data acquisition in longitudinal studies necessitates optimal pipelines for measuring change.
- Identifying the most effective quantification strategy is crucial for advancing Alzheimer's disease (AD) research.
Purpose of the Study:
- To compare 415 distinct tau PET quantification methods for their impact on clinical group differences in annual Standardized Uptake Value Ratio (SUVR) change.
- To evaluate the longitudinal measurement repeatability of these methods using linear mixed-effects models.
- To determine the optimal pipeline for accurate and repeatable tau PET analysis in Alzheimer's disease.
Main Methods:
- Analysis of MRI and Flortaucipir PET scans from 97 participants (cognitively unimpaired or AD pathology).
- Comparison of cross-sectional and longitudinal pipelines (FreeSurfer 6.0, SPM12-based) with variations in target regions, partial volume correction (PVC), reference regions, masking, and statistics.
- Evaluation based on differences in annual SUVR change between clinical groups and measurement repeatability (error term from linear mixed-effects model).
Main Results:
- Larger temporal lobe regions of interest (ROIs) significantly outperformed the entorhinal cortex, reducing estimated clinical trial sample sizes.
- Longitudinal processing pipelines demonstrated superior performance over cross-sectional approaches in terms of repeatability and efficiency.
- Specific reference regions, including supratentorial white matter, yielded better results than traditional cerebellar or pontine regions.
Conclusions:
- Longitudinal SUVR methods combined with a temporal lobe meta-ROI, a composite reference region, 2-class partial volume correction, and median statistics are favored.
- Optimized quantification pipelines enhance the sensitivity to detect clinical group differences and improve the precision of longitudinal tau PET measurements.
- These findings provide critical guidance for selecting robust tau PET quantification strategies in Alzheimer's disease research and clinical trials.
