PSEG Workflow > ORGAN SEGMENTATION

Background

Preclinical imaging systems allow following the uptake and metabolism of labeled compounds in whole mice and major parts of rats. To exploit this wealth of information with regard to phamacokinetics, it is necessary to localize all relevant tissue structures and assess their tracer concentration in time, which is then analyzed with kinetic models.

Manual segmentation is a tedious process and prone to considerable subjective variability. It is challenging for studies involving many animals, has a direct impact on the accuracy of the study endpoint and consequently on the required sample size.

Solution Overview

The ORGAN SEPARATION approach in the PSEG tool supports the semi-automatic segmentation of dynamic rodent PET studies within only a few minutes. It applies a patented clustering approach for grouping neighboring image pixels with similar kinetic behaviors. This process creates a hierarchy of compact tissue segments covering the whole body. Segments on low hierarchy levels are small and highly homogeneous regarding kinetics, whereas they grow towards higher levels and become kinetically more variable. The user can interactively browse through the hierarchy levels and assign an organ label when an appropriate representation has been found. If tissue structures cannot directly be represented by a segment or a union of segments, the user can interactively outline them in an anatomical reference image.

Once all relevant tissue structures have been segmented, PSEG readily calculates their average PET uptake curves (also time-activity curves, TAC) and optionally corrects them for the partial-volume effect.

Segmentation Methodology

PSEG implements the local means analysis (LMA) method licensed from CEA, Orsay, France. This method aims at segmenting "functional organs" characterized by a particular pharmacokinetics. A functional organ only corresponds to an entire anatomical organ if the organ is functionally homogeneous. Otherwise, the functionally of different organ parts are separated and can be treated individually. The patented LMA method has been shown to be robust regarding the low signal-to-noise ratio, the limited spatial resolution and potential organ movement of dynamic rodent PET studies [3,4,5].

The LMA segmentation includes the following processing steps:

▪Exclusion of the noisy background from segmentation.

▪Automated localization of pixels in the organ centers.

▪Computation of the local pharmacokinetics and global noise parameters.

▪Parcellation of the body volume in the image into a predefined number of regions.

▪Hierarchical fusion of the regions to form increasingly larger, connected regions. On each level of the hierarchy, the whole body is covered by disjoint regions.

Tissue structures which cannot be isolated by the LMA segmentation because they do not kinetically differ from the neighborhood need to be outlined using the general VOI features of PMOD.

The following sections describe the stepwise workflow for performing an ORGAN SEPARATION segmentation. For details of the common parts data loading and masking, the user is referred to the corresponding sections of the PERCIST segmentation.