Kinetic and Wavelet Analysis of Dynamic FDG PET Data in Human Glioblastoma

Brain tumor detection and therapeutic evaluation continue to be very challenging. Distinguishing between radiation induced necrosis and recurrent or viable residual tumor in human glioblastoma has proved to be a particularly difficult task. The aim of this work is to extract the time course features of the pharmacokinetics of F18-FDG after intra-venous injection using dynamic Positron Emission Tomography (PET) imaging and utilize the features based on kinetic modeling and wavelet transform (WT) as new methods for brain tumor diagnosis other than MRI and static PET images. These methods provide additional functional information to assist with classification of tumor types, i.e., metastasis, recurrent tumor, versus treatment induced changes, such as radiation necrosis. When combined with the novel dedicated brain PET imagers this dynamic PET technique can provide an even more powerful new diagnostic tool.