Development and evaluation of image preprocessing pipelines for the Centiloid method on Down Syndrome data.

Area of Science:

• Neuroimaging
• Biomedical Engineering
• Neurology

Background:

• Centiloid is a standardized method for quantifying brain amyloid using MRI and PET scans.
• The standard Centiloid process has a high failure rate in Down syndrome (DS) subjects.
• Improving Centiloid processing success in DS is crucial for neurodegenerative disease research.

Purpose of the Study:

• To evaluate different imaging preprocessing methods (PMs) to enhance the success rate of Centiloid processing in DS subjects.
• To identify optimal preprocessing pipelines for DS populations.
• To ensure Centiloid results adhere to established standards.

Main Methods:

• Preprocessing methods were developed by combining image origin reset, filtering, MRI bias correction, and skull stripping.
• The performance of these PMs was evaluated using The Global Alzheimer's Association Interactive Network dataset for standard adherence.
• PMs were specifically assessed for their suitability in the DS population using the NiAD dataset.

Main Results:

• Five preprocessing pipelines were identified as effective in improving Centiloid processing success.
• The success rate for Centiloid processing in the DS cohort increased from 61.3% to 95.6% with the optimized PMs.
• The developed preprocessing strategies demonstrated substantial improvements in the yield of usable imaging data.

Conclusions:

• The study proposes a robust image preprocessing pipeline for Centiloid analysis in DS.
• These optimized preprocessing steps significantly enhance the success rate of Centiloid processing in DS.
• The findings facilitate more reliable amyloid quantification in DS individuals, aiding research and clinical applications.