PET images for reports

Original PET image files contain quantitative volumetric data in formats dedicated to nuclear imaging and PET, and these files are not readable by common office or photo editing software. Visualization of volumetric data requires the use of tailored display algorithms such as direct volume rendering, maximum intensity projection, or stereotactic surface projection.

There are several commercial and non-commercial PET image analysis tools, including CarimasTM, which will display PET images on screen, and allow you to save the displayed image in common raster file formats like TIFF, GIF, JPEG or PNG; or you can use the Print Screen function of your operating system to save the screen-shot image for further editing.

Comparison of PET images

For reporting purposes usually one or more images need to be compared, and then it is important that the images are displayed in the same colour scale. PET image analysis software usually allow user to set the maximum and minimum pixel values when displaying images.

Although original PET images contain quantitative pixel values, for example in units Bq/cc, these are not directly comparable between studies because tissue uptake is affected by injected dose, subject weight, and plasma clearance. These should be accounted for by computing parametric images, at least SUV images.

Presentation images using scripts

Simple 2D TIFF images without any volume raycasting can be made using command-line tool img2tif, which may come handy when lots of images need to be generated for reports or quality control. Colour scale can be selected using appropriate options. Most photo editing programs can read TIFF format, and these files can be directly drag-and-dropped into PowerPoint. For setting the colour scale, the maximum pixel value of all images is needed and program imgmax can be used for that.

TIFF images can be converted to other (smaller) raster file formats, for example using ImageMagick. ImageMagick is very powerful tool to do also other formatting and editing for images in scripts. For example, the following command both converts the image from TIFF format to PNG format and adds the study number as a title to image:

convert -font Times-New-Roman-Bold -pointsize 36 -gravity center label:ia456 -append ia456suv.tif ia456suv.png

See also:


Fujiwara T, Miyake M, Watanuki S, Media MA, Itoh M, Fukuda H. Easy detection of tumor in oncologic whole-body PET by projection reconstruction images with maximum intensity projection algorithm. Ann Nucl Med. 1999; 13(3): 199-203.

Kim J, Cai TW, Fulham M, Eberl S, Feng DD. Data visualization and display. In: Feng DD (ed.): Biomedical Information Technology. Elsevier, 2008, pp 211-227.

Lawonn K, Smit N, Preim B, Vilanova A. Eurographics Workshop on Visual Computing for Biology and Medicine, 2015. doi: 10.2312/vcbm.20151213.

Lilja J, Thurfjell L, Sörensen J. Visualization and quantification of 3-dimensional stereotactic surface projections for 18F-Flutemetamol PET using variable depth. J Nucl Med. 2016; 57(7): 1078-1083. doi: 10.2967/jnumed.115.169169.

Miller TR, Starren JB, Grothe RA Jr. Three-dimensional display of positron emission tomography of the heart. J Nucl Med. 1988; 29(4): 530-537.

Stytz MR, Frieder G, Frieder O. Three-dimensional medical imaging: algorithms and computer systems. ACM Computing Surveys 1991; 23(4): 421-499.

Wallis JW, Miller TR. Three-dimensional display in nuclear medicine and radiology. J Nucl Med. 1991; 32(3): 534-546.


Updated at: 2018-01-21
Created at: 2005-05-24
Written by: Vesa Oikonen