Related Experiment Video
Updated: Jan 11, 2026
PET and MRI Guided Irradiation of a Glioblastoma Rat Model Using a Micro-irradiator
Published on: December 28, 2017
AI-driven parametric FDG brain PET and brain tumor segmentation for multi-modal convolution neural network to
Bijoy Kundu1, Zoraiz Qureshi2, Rugved Chavan2
1Departments of Radiology and Medical Imaging, University of Virginia, Charlottesville, VA, United States; Department of Biomedical Engineering, University of Virginia, Charlottesville, VA, United States.
Abstract:
Glioblastoma (GBM) accounts for 52 % of all malignant primary brain tumors. The main treatment regimens include surgical resection followed by chemoradiotherapy. Despite these, the median survival of patients with GBM is only 15 months. To make things worse the neuroimaging characteristics mimic tumor recurrence. Traditional static Fluorine-18 fluorodeoxyglucose (FDG) positron emission tomography (PET) is also unreliable. In this review manuscript we describe advanced PET methods to differentiate tumor progression (TP) from treatment related necrosis (TN) in post treated GBM patients. Dynamic FDG PET (dPET), an advance from traditional static FDG PET, may prove advantageous in clinical staging. Quantifying dPET data is however challenging. In this review we showcase new work for an end-to-end novel AI platform for automated blood input computation to quantify dPET data. Next, we review new work on a large AI model for multimodal brain tumor segmentation from post-treated BraTS2024 MRI datasets and finally a multi-modal AI platform including dPET and MRI for classification of TP vs TN.
