Presurgical Mapping of Brain Tumors Using Statistical Probabilistic Anatomical Maps

Abstract

Purpose: Multi-tracer neuroimaging is widely used for the diagnosis, radiotherapy, and neurosurgery of brain gliomas. Anatomical and functional information is important to avoid having neurological deficits induced during the resection or radiation therapy of brain gliomas. The aim of this study was to investigate presurgical anatomical labeling of brain gliomas on FLT and FET using statistical probabilistic anatomic maps (SPAM), which are images of cerebral cortical, cerebellar, and subcortical volumes of interest (VOIs). Methods: FDG, FLT, and FET PET scans were acquired. FLT and FET PET images were coregistered to the FDG PET images, which were then spatially normalized onto the target brain. An inverse spatial normalization parameter was calculated and applied to SPAM. For the anatomical labeling of brain glioma regions, the volumes of brain gliomason FLT and FET images were extracted using segmentation. Probabilistic information of the glioma region was then calculated using SPAM and the segmented glioma volumes. SPM and an in-house program were used for image processing. Results: The probability of SPAM labeling a brain glioma region could be extracted using the inverse normalized SPAM and segmented glioma regions. In a sample case, the probabilistic anatomical region of the glioma included 21% of the postcentral gyrus, 12% of the superior parietal gyrus, and 6% of the angular gyrus. Conclusion: Anatomical information about brain gliomas could be extracted using SPAM. This proposed method would be optional for presurgical mapping, to avoid an additional functional mapping study that might otherwise be necessary to avoid producing neurological deficits.

FDG Whole-Body PET/MRI in Oncology: a Systematic Review

“…FDG PETcould suggest a 95 % of diagnostic accuracy for detecting functional activation zone. FDG uptake of brain tumor is related with malignancygrade,anddirect comparisonof PETand MRI images provided additive diagnostic information such as tumor location, boundary, and heterogeneity [12][13][14].…”