A High-Performance Instrument for Brain Imaging in Combination with Whole Body PET-CT for Oncology

About This Grant

A High-Performance Instrument for Dedicated Brain Imaging and in Combination with Whole Body PET for Oncology: The goal of this project is to develop a high-performance brain PET scanner in an academic- industry partnership with Siemens to operate in tandem with a total-body PET that will open up new oncologic applications that leverage simultaneous high-resolution brain and body imaging, and also be capable to operate as a stand-alone device for brain-only imaging to widen its clinical use. Dynamic brain-body imaging is currently only provided by total-body PET systems, though the spatial resolution of these devices is limited for brain imaging. The new instrument will provide < 2-mm spatial resolution (isotropic) and system TOF resolution near 160 ps for an unprecedented combination of spatial resolution and effective sensitivity. We will achieve our goal through 3 specific aims: (1) Develop and test performance of the brain scanner, and integrated with TB PET; (2) Adapt and evaluate image generation methods for quantitative brain imaging; and (3) Test clinical performance and evaluate brain-body imaging protocols in humans with emerging tracers, including 18F-FES and 18F-Fluciclovine (FACBC), as well as 18F-FDG. In Aim 1 we will develop and optimize a high-resolution detector design based on Siemens state-of-the-art technology, using experimental measurements as input to Monte Carlo simulations to model the system and test the data correction and image generation methods developed in Aim 2. The scanner development will be driven by requirements for high performance, using robust manufacturing practices. The scanner will be built and integrated with the PennPET Explorer scanner and performance will be characterized using physical measures and phantom studies. In Aim 2 we will develop and apply methodology to produce accurate quantitative images focusing on data driven methodologies for calibrations and for brain motion correction, energy-based methods for scatter correction, deep learning-based methods for attenuation correction (for use without CT), and image- based resolution modeling to achieve uniform spatial resolution within the imaging FOV. In Aim 3, we will design human protocols and evaluate the clinical performance in two small cohorts of patients with melanoma, breast and lung cancer undergoing Amino Acid (FACBC) and FDG PET or with breast cancer starting estrogen receptor (ER)-targeted therapy and undergoing 18F-FES and FDG PET. These studies will demonstrate the benefits of the high-resolution brain scanner and its capability to provide simultaneous brain-body imaging for oncologic applications. The human studies with the Brain-TB PET scanner will provide feedback on the technology, as well as the patient and end-user experience that will guide translation of the prototype to a commercial product integrated with the Siemens Vision Quadra.