Centiloid Scale Standardization for Dementia Research
- ceva130
- 3 days ago
- 2 min read

Positron Emission Tomography (PET) imaging of amyloid‑β plaques plays a central role in Alzheimer’s disease (AD) research. However, variability across tracers, scanners, and processing pipelines can make results difficult to compare between studies. The Centiloid (CL) scale was developed to address this issue by providing a standardized, tracer‑independent framework for quantifying amyloid burden.
What is the Centiloid Scale?
The Centiloid scale expresses amyloid PET results on a universal metric:
0 CL: typical amyloid level in young, healthy controls
100 CL: typical amyloid level in patients with mild-to-moderate Alzheimer’s disease
This linear scaling allows values from different tracers (e.g., PiB, florbetapir, flutemetamol) to be directly compared.
Basic Workflow for Centiloid Standardization
Acquire PET Data
Obtain amyloid PET scans using your tracer of choice.
Ensure consistent acquisition protocols where possible.
Preprocessing
Perform motion correction, spatial normalization (typically to MNI space), and smoothing.
Co-register PET images with structural MRI if available.
Define Regions of Interest (ROIs)
Use the standard Centiloid cortical target region and reference region (often whole cerebellum).
Apply validated ROI masks provided by the Centiloid project.
Calculate SUVR
Compute the Standard Uptake Value Ratio (SUVR): SUVR=target region uptakereference region uptakeSUVR = \frac{\text{target region uptake}}{\text{reference region uptake}}SUVR=reference region uptaketarget region uptake
Convert SUVR to Centiloids
Apply a linear transformation derived from calibration studies: CL=a×SUVR+bCL = a \times SUVR + bCL=a×SUVR+b
The coefficients (a, b) are tracer- and pipeline-specific and must be established by cross-calibration with the Centiloid reference method (usually PiB).
Quality Control
Verify preprocessing accuracy and ROI alignment.
Check for outliers or artifacts.
Why Use Centiloids?
Comparability: Enables direct comparison across tracers, sites, and studies
Reproducibility: Reduces methodological variability
Clinical relevance: Facilitates threshold definition for amyloid positivity
Longitudinal consistency: Improves tracking of disease progression and treatment effects
Key Considerations
Always validate your processing pipeline against the standard Centiloid protocol.
Use published conversion equations or derive your own with appropriate datasets.
Report both SUVR and Centiloid values when possible for transparency.


