Abstract
Availability of input and organ functions is a prerequisite for analysis of dynamic image sequences in scintigraphy and Positron Emission Tomography (PET) via kinetic models. This task is typically done manually by a human operator who may be unreliable.
We propose a probabilistic model based on physiological assumption that Time-Activity Curves (TACs) arise as a convolution of an input function and organ-specific kernels. The model is solved via the Variational Bayes estimation procedure and provides estimates of the organ images, the TACs, and the input function as results.
The ability of the resulting algorithm to extract the input function is tested on data from dynamic renal scintigraphy. The estimated input function was compared with the common estimate based on manual selection of the heart ROI.
The method was applied to the problem of relative renal function estimation and the results are compared with competing techniques. Results of comparison on a dataset of 99 patients demonstrate usefulness of the proposed method.