Utilizing the Centiloid scale in cross-sectional and longitudinal PiB PET studies

Area of Science:

• Neuroimaging
• Biomarkers
• Dementing Disorders

Background:

• Amyloid imaging is crucial for dementia research and diagnosis.
• Lack of standardized quantification methods hinders effective use of amyloid imaging data.
• The Centiloid approach was developed to standardize amyloid imaging quantification.

Purpose of the Study:

• To implement and evaluate the Centiloid approach for amyloid imaging quantification.
• To assess the impact of varying image analysis methodologies on Centiloid values.
• To determine the comparability of Centiloid scales across different analytical pipelines.

Main Methods:

• Implementation of the Centiloid approach for amyloid imaging data.
• Analysis of cross-sectional and longitudinal datasets.
• Comparison of Centiloid values derived from different image analysis techniques.

Main Results:

• The Centiloid approach successfully standardizes amyloid burden measurements onto a common scale (CL).
• Statistically significant differences (p < 0.05) in variability and group means were observed due to underlying analytical approaches.
• Thresholds for amyloid positivity (5.7-11.9 CL) and reliable worsening (-2.0-11.0 CL) varied significantly across methods.

Conclusions:

• While the Centiloid approach provides a common scale, variations in image analysis persist.
• Differences in acquisition protocols and analysis techniques can still impact combined multi-center analyses.
• Careful consideration of analytical methodologies is necessary when interpreting Centiloid values from different sources.