Related Experiment Video
Updated: Jan 15, 2026
A Semi-high-throughput Imaging Method and Data Visualization Toolkit to Analyze C. elegans Embryonic Development
Published on: October 29, 2019
Development and evaluation of image preprocessing pipelines for the Centiloid method on Down Syndrome data
Weiquan Luo1, Davneet S Minhas2, Ethan D Rubenstein3
1Department of Bioengineering, University of Pittsburgh, Swanson School of Engineering, Pittsburgh, Pennsylvania, USA.
Background:
Centiloid provides a standardized process to quantify brain amyloid in which a subject's T1 magnetic resonance imaging (MRI) and amyloid positron emission tomography (PET) scans are registered and warped to Montreal Neurological Institute 152 space using prescribed procedures. The method has a high failure rate in Down syndrome (DS) subjects from the Neurodegeneration in Aging Down Syndrome (NiAD) project. We evaluate imaging preprocessing methods (PMs) to improve the DS success rate.
Methods:
PMs were constructed from combinations of image origin reset, filtering, MRI bias correction, and MRI skull stripping. Centiloid results were evaluated for adherence to standards using The Global Alzheimer's Association Interactive Network dataset. PMs were also evaluated using the NiAD dataset to judge their suitability for the DS population. DS PM evaluation procedures were developed corresponding to those specified for non-DS populations.
Results:
Five accepted PMs improved the Centiloid-processing success rate in the DS cohort from 61.3% to 95.6%.
Discussion:
The identified combinations of preprocessing steps substantially improved the success rate of Centiloid processing in DS.
Highlights:
Image preprocessing pipeline is proposed for Centiloid analysis of DS. Preprocessing pipelines are evaluated for adherence to Centiloid standards. Pipelines are evaluated for improvement in yield of usable imaging data. Preprocessing of amyloid imaging data resulted in a large yield improvement.
