Effect of improved spatial resolution on PET SUVR estimates in the LEADS study of sporadic early-onset Alzheimer's disease
Schonhaut, D. R.; La Joie, R.
Show abstract
BackgroundThe Longitudinal Early-onset Alzheimers Disease Study (LEADS) recently transitioned from processing positron emission tomography (PET) scans at 8mm to 6mm resolution in early 2024. Concurrent with this improved spatial resolution, we revised our processing methods for defining reference regions used in amyloid and tau PET analyses, with the goal to maintain alignment with MRI-based PET processing in the Alzheimers Disease Neuroimaging Initiative (ADNI). MethodsWe describe the following changes to the LEADS MRI-based PET processing pipeline: O_LITransition from 8mm to 6mm PET scans. C_LIO_LIRevised method to calculate a composite reference region scaling factor for longitudinal analysis of 18F-Florbetaben (FBB). C_LIO_LIRevised method to define the eroded subcortical white matter region, which is used in longitudinal analyses of both FBB and 18F-Flortaucipir (FTP). C_LIO_LICorrected method to define the inferior cerebellar gray matter reference region used in cross-sectional analysis of FTP. C_LI We evaluated these changes by comparing 8mm standardized uptake value ratios (SUVRs) processed with the original LEADS PET pipeline to 6mm SUVRs processed with the updated pipeline. This analysis was conducted in 564 FBB, 555 FTP, and 134 FDG scans acquired in LEADS participants at baseline between May 2018 and September 2023. ResultsWe found near-perfect correlations across all tracers and reference regions (R2 > 0.99). However, 6mm SUVRs had greater dynamic range than 8mm SUVRs, with 8mm-to-6mm linear transformation slopes ranging from 1.08 to 1.22. ConclusionsOur findings closely parallel ADNIs experience transitioning to higher resolution PET. Amyloid PET Centiloid conversion equations, which were originally derived using 8mm data, should be re-estimated using 6mm data to maintain optimal harmonization. LEADS data users should not mix 6mm and 8mm PET data and should instead exclusively use 6mm PET data going forward, as these measures better reflect modern scanner capabilities while maintaining compatibility with ADNI. LEADS data users who are unsure of the origin or resolution of their PET values are encouraged to contact us for advice at LEADS.PETCORE@ucsf.edu.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Automated Long Axial Field of View PET Image Processing and Kinetic Modelling with the TurBO Toolbox 94%
- Validation of cardiac image derived input functions for functional PET quantification 93%
- Non-invasive assessment of stimulation-specific changes in cerebral glucose metabolism with functional PET 92%
Similar papers in this journal
- Evaluation of ComBat harmonization for reducing across-tracer biases in regional amyloid PET analyses 95%
- OpenMAP-T1: A Rapid Deep Learning Approach to Parcellate 280 Anatomical Regions to Cover the Whole Brain 94%
- Comparison of structural MRI brain measures between 1.5T and 3T: data from the Lothian Birth Cohort 1936 93%
Similar papers in this journal
- Reliability Assessment of Tissue Classification Algorithms for Multi-Center and Multi-Scanner Data 93%
- Automated joint skull-stripping and segmentation with Multi-Task U-Net in large mouse brain MRI databases 93%
- Fusion of quantitative susceptibility maps and T1-weighted images improve braintissue contrast in primates 93%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.