|
|
|
|
|
Step-wise processing is started by selecting the Segment tab.
PET Image Loading
The Load PET button for loading the PET images is located in the right control area. It behaves as an option button which needs to be set to the appropriate data format with the indicated arrow. For loading images which are not saved in a PMOD database it is recommended to apply the Autodetect format which will figure out the appropriate image loader.
PSEG assumes that the loaded PET series is dynamic. It generates a new series by averaging a range of frames and assigns it to the AVGERAGED tab. The averaging range can be defined by the From and To number fields, or dragging the range indicators in the Aver bar. After any modification of the range, the average is recalculated and the display updated. The aim of the averaging is to generate an image which is useful as an anatomical reference for the segment interpretation, as well as for creating the body mask. The original and the averaged images are shown in a fusion rendering which can be adjusted in the fusion control area below the tabs for selecting the individual images.
PET Image Cropping
If the PET field-of-view is much larger than the body of the animal, the data set should be reduced in order to save RAM and optimize processing. To do so, please enable the Crop box and place the center of the yellow crop volume indication by clicking into the image. The edge size in [mm] of the box can be adjusted for each direction by selecting the size in the corresponding list, or by selecting the Size button and entering the three edge lengths. The 
button initiates cropping, whereby the original data are replaced. If cropping is not activated manually, a request will be shown when proceeding to the next step. Note: The cropping operation is only allowed once.
Body Mask Definition
The segmentation will operate on all pixels contained in a mask which is usually derived from the data. Please set the Species selection according to the studied subject. It establishes proper defaults for the morphological operations which are later applied to the data. The Mask selection allows switching between three methods of mask derivation: PET, CT and MR.
PET-Based Body Mask
With PET masking the body mask is derived from the averaged PET series. There are five masking approaches available:
In addition to the procedure selected above, morphological opening/closing operations will be applied to remove isolated pixels and fill holes.
Please select a definition appropriate for your PET series. If there are no matched MR or CT images, proceed with the Mask action button, otherwise proceed to the additional image loading with the +CT | MR button.
MR- or CT-Based Body Mask
If matched CT or MR series of the same animal are available, they can not only be used as reference images for the segmentation assignment, but also for creation of the body mask. In order to apply MR- or CT-based masking please set the Mask choice accordingly. The change is reflected in the parameter area as illustrated below.
The PET masking options are no longer available, and the action buttons are replaced by a single button for progressing to the corresponding image loading page.