Precise knowledge of the functional brain areas in the individual subject is crucial for the accurate quantitative analysis of brain PET images and their meaningful interpretation. Often however, functional regions are manually outlined in a casual manner, with subjective results as the outcome. For rodent brain images, this fundamental flaw can be overcome with PMOD’s PNROD tool which offers objective region outlining via easy step-by-step procedures.
Automatic Rodent Brain Regions by Atlas Adjustment
A well established way of obtaining brain region outlines automatically is by adjusting brain atlases to a subject brain. This is achieved by calculating a spatial transformation which maps the brain anatomy represented in the atlas to the actual brain anatomy of the subject. The transformation is called "spatial normalization", and correspondingly its estimation step "normalization". Various types of normalizations are supported in PNROD: elastic procedures which warp the entire anatomy, affine matching which only scales and shears the brain as a whole, and rigid matching which assumes that the imaged animal brain is practically identical to the atlas brain. If all automatic procedures fail, rigid matching can be manually performed as a last resort. Naturally, the estimation of the atlas transformation is crucial for the whole analysis.
Once a successful normalization has been found, the spatial transformation between the anatomies can be used to project the brain areas encoded in the atlas to the subject brain. The user should always inspect the adequacy of the resulting outlines (called VOIs i.e. Volumes-of-Interest) and has tools available to adjust them locally.
Finally, statistics can be calculated within the brain regions resulting in signal average, volume, and other statistics. In case of dynamic acquisitions, the average regional signal is conveniently obtained as a time curve and can directly be transferred to the kinetic modeling tool (PKIN) for absolute quantification of tissue function.
Parametric Mapping
If the pixel-wise modeling tool PXMOD has been licensed, PNROD also embeds the PXMOD Parametric Mapping workflow which can be run after regional outlining has been completed. This integration has several advantages: (1) The regions generated by PNROD can be used during the PXMOD model configuration. (2) The resulting parametric maps can immediately be regionally analyzed. (3) The parametric maps can be calculated in the subject or the atlas space.
Workflows
PNROD supports two different workflows:
1.Dual-image workflow with two images from the same subject brain: In such studies, the target image for analysis is typically a functional (PET, SPECT), and an anatomical image (MR, CT) serves as the anatomical backdrop and is normally used for the spatial normalization. In hybrid studies, the two images are inherently aligned. If the two brain images were acquired in separate scans, an additional rigid matching step is required to bring them into alignment. Once the transformations have been determined, each image can be transformed into the space of any other image or the atlas. Similarly, the atlas regions can be transformed to any of the images. This has the advantage that the user can calculate the regional statistics in the original functional image, in the image transformed to the anatomical image space, or in the image transformed to the atlas space. It is .
2.Single image workflow: If only a single image is available, the transformation to the atlas anatomy has to be determined directly. In this situation, statistics can be calculated in the original image or in the image transformed to the atlas space.
Starting the PNROD Tool
The PNROD tool is started with the Rodent Brain button
from the PMOD ToolBox or by directly dragging image files onto the above button.