Development of the Human Dynamic Neurochemical Connectome Scanner

21.09.2020 bis 31.05.2025

We seek support to develop and build the next generation 7-Tesla magnetic resonance (MR)-compatible positron emission tomography (PET) brain scanner with dramatically improved spatiotemporal resolution (HSTR-BrainPET). PET and MRI are two of the most powerful imaging modalities currently in use for studying the human brain. Recently, scanners capable of simultaneous PET and MR whole-body data acquisition in human subjects have become commercially available. However, there is no equivalent dedicated head device on the market to address the needs of the researchers and clinicians focusing on the brain and the performance of whole-body devices is rather limited for this purpose. More importantly, although current PET technology achieves high molecular sensitivity with a broad set of probes for neurochemical targets, PET still lacks the capability to track dynamic changes in a time scale comparable to functional processes. Our main goal is to build an MR-compatible PET camera with very high sensitivity to enable truly dynamic PET imaging of brain neurotransmission. One of the first MR-compatible brain PET prototypes was installed at the Martinos Center in 2008 when human PET/MR imaging was in its infancy. Following a close collaboration with Siemens to address the remaining technical challenges, proof-of-principle PET/MR studies demonstrating the advantages and potential of this novel imaging modality were performed. A decade later, a new type of photon detector technology has reached a level of maturity that would allow us to build the next generation integrated system with dramatically improved spatiotemporal resolution. We propose to address the hardware and software challenges in building 7-T MR-compatible PET technology purpose-built to extend the temporal window of PET down to just a few seconds. Additionally, the substantial improvement in spatial resolution will also allow for imaging of cortical substructures and nuclei that cannot be resolved with current state-of-the-art devices. Specifically, we propose to: (1) Build the hardware components of the HSTR-BrainPET insert, integrate it with the 7-T MR scanner and characterize the combined device; (2) Implement the software for PET data acquisition, processing and image reconstruction adapted to the non-conventional geometry we are proposing; (3) Apply the integrated scanner to dynamic assessment of neurochemical events and brain activation in healthy human subjects.