RT Journal Article
SR Electronic
T1 Automated quantification of PET lesions using anatomical information in CT images
JF Journal of Nuclear Medicine
JO J Nucl Med
FD Society of Nuclear Medicine
SP 1036
OP 1036
VO 50
IS supplement 2
A1 Demirkaya, Omer
A1 Abouzied, Mohei
YR 2009
UL http://jnm.snmjournals.org/content/50/supplement_2/1036.abstract
AB 1036 Learning Objectives 1. Registration and sampling of both PET and CT images. 2. Segmentation of PET and CT images. 3. Volume of interest definitions and Quantification. Summary: The flow chart of the steps that were used in our method is shown in Figure 1. As seen in the flow chart, the first step is to register and resample PET and CT images so that they lie in a common coordinate system and have the same pixel size. Then the bone structures were segmented using simple thresholding. After the extraction of the bone, the maximum standard uptake values (SUV) in the bone were computed for each slice. The maximum SUV function (of slice location) was analyzed to find the suspected slices, that is, the slice that may have bone metastases. It is expect that the suspected slices have significantly larger maximum SUV values than the other slices in general. The suspected slices were subjected to the visual confirmation by a nuclear medicine physician experienced in PET/CT who will also describe the structural changes in the bone CT window whether lytic, sclerotic or mixed type. Then, the chosen slice will be subject to further region-of-interest (ROI) and volume-of-interest (VOI) analysis in both PET and CT images to calculate maximum and average SUV values as well as the average CT value within the VOIs.