An activity-subspace approach for estimating the integrated input function and relative distribution volume in PET...

Area of Science:

• Nuclear Medicine
• Radiochemistry
• Neuroimaging

Background:

• Dynamic positron emission tomography (PET) enables quantification of physiological and biochemical parameters by measuring neuroreceptor distributions.
• Accurate neuroreceptor concentration quantification is crucial for understanding brain function and disease.
• Current compartment modeling methods rely on arterial blood input functions or reference regions, posing limitations.

Purpose of the Study:

• To develop a novel method for estimating the input function in dynamic PET imaging, obviating the need for arterial blood measurements or a reference region.
• To enable accurate, image-wide quantification of neuroreceptor concentrations and physiological parameters.
• To refine the estimation of the distribution volume (DV) parameter iteratively.

Main Methods:

• Proposed a novel concept of activity subspace for input function estimation.
• Estimated the input function through the analysis of the intersection of activity subspaces.
• Iteratively refined the input function and distribution volume (DV) parameter to obtain a parametric image of total DV.

Main Results:

• The proposed method demonstrated superior performance compared to the blind estimation method, iterative quadratic maximum-likelihood (IQML), in simulations.
• The method successfully estimated the input function and distribution volume (DV) parameter.
• Evaluation on a brain PET dataset confirmed the method's efficacy.

Conclusions:

• The novel activity subspace analysis provides an effective approach for input function estimation in dynamic PET.
• This method enhances the accuracy and feasibility of neuroreceptor binding studies by removing the requirement for invasive arterial sampling or reference regions.
• The technique facilitates image-wide quantification of physiological and biochemical parameters in PET neuroimaging.