Clustering-initiated factor analysis application for tissue classification in dynamic brain positron emission...

Area of Science:

• Neuroscience
• Medical Imaging
• Biophysics

Background:

• Dynamic brain positron emission tomography (PET) analysis often uses region of interest methods with approximate kinetic models.
• These standard methods can limit the accuracy and reproducibility of quantifying specific tracer binding in brain tissues.

Purpose of the Study:

• To introduce and validate a novel dynamic image processing technique, clustering-initiated factor analysis (CIFA).
• To accurately quantify the fraction of tissues exhibiting specific tracer binding in dynamic brain PET studies.

Main Methods:

• Developed and applied clustering-initiated factor analysis (CIFA) for dynamic PET image processing.
• Utilized CIFA to determine time-activity curves and spatial distributions of radiotracer concentration.
• Analyzed PET images acquired with (11)C-Pittsburgh Compound B, a tracer for β-amyloid.

Main Results:

• CIFA accurately determined time-activity curves and spatial distributions, including the arterial input function.
• The fraction of specific binding tissues quantified by CIFA correlated well with Logan graphical analysis.
• Demonstrated CIFA's capability to extract arterial and venous blood curves from PET images.

Conclusions:

• Clustering-initiated factor analysis (CIFA) offers an accurate and convenient tool for analyzing tracer kinetics and measuring specific binding tissue concentration in dynamic PET.
• CIFA enhances the analysis of various PET tracers and can function effectively even with limited temporal resolution.