Image data is stored in a file on computer disk, or database. In an image data file:
- Radioactivity concentration is assigned to each image pixel
- Each time “frame” has its own image
File contains these radioactivity concentration values, stored in a very long array. File also contains metainformation (“header”) about the size of image, size of image pixels, scan time, etc. Image file:
- Image data can be stored in several quantitative image file formats (DICOM, NIfTI, Interfile, ECAT, Analyze, etc); scanner manufacturers have their own image file formats or DICOM versions.
- Dedicated software necessary to process the PET image files. Both commercial and free, even open source, tools are available, for example Carimas, SPM, Amide, and PMOD. PET scanners are delivered with basic image analysis tools.
- Common image formats (JPEG, TIFF, GIF, etc) can be generated from PET image files, but those are for viewing only — quantitative information is lost in these image formats.
“Static image” contains only one time frame, usually scanned a certain time after tracer injection (late-scan). These data can be analyzed visually, or used to calculate for example SUV, FUR, or ratio images.
[15O]H2O PET studies contain often only static imaging, starting from the tracer injection. Static images can be used in brain activation studies, or perfusion can be quantified with ARG method.
How to find the PET images in TPC?
Please read the SOP7539.
Mask image is an image, which is related to a certain PET image, but which does not contain quantitative concentration or parameter values, but instead each voxel has it’s own label (0, 1, 2, …). Mask images can be used to store a selection of image voxels, for example when defining ROIs, or when masking out certain image areas.
- Late scan image
- Catenation of PET images
- Parametric image
- Mask image
- Decay correction
- Partial volume effect
- Presentation images
- Units of PET data
- Tools for processing image files
- Model calculations for PET images
- Image formats
- Image reconstruction
Updated at: 2018-02-11
Created at: 2014-02-06
Written by: Vesa Oikonen