Real-Time Motion Correction Method for a Portable, High-Resolution TOF PET Scanner for Imaging the Brain, Head, and Neck

About This Grant

Head and neck cancer (HNC), including cancers of the oral cavity and oropharynx, is the sixth most common cancer and the ninth leading cause of cancer-related deaths globally. Despite advances in PET imaging, a critical tool in staging and treatment planning, the detection of cervical lymph node metastases remains challenging due to the limited resolution of conventional whole-body PET scanners. This limitation results in high false-negative rates, particularly for small (<7 mm) metastases in clinically node-negative patients. Due to the high prevalence (20-30%) of occult lymph node metastases, many HNC patients undergo elective neck dissections to remove potentially affected nodes, despite the associated risks and morbidity. Addressing this unmet need, the Head PETcoil is a portable, high-resolution, time-of-flight (TOF) PET system designed for brain, head, and neck imaging. Compared to conventional PET systems, the Head PETcoil offers nearly six times finer spatial resolution and three times higher photon sensitivity, improving the detection of tumors and small lymph node metastasis. To realize this improved resolution in patient studies, it is critical to correct for image blurring due to head motion, which can cause displacements that range from 2 to 5 mm. This Phase I project aims to develop and validate real-time motion correction software for the Head PETcoil. Aim 1 is to develop ultra-fast data processing and image reconstruction software capable of generating short timeframe (1-5 seconds) PET images in real time during data acquisition. Aim 2 is to create real-time motion estimation and correction software to calculate motion from short timeframe images and produce motion-corrected images within 5 minutes after scan completion. Aim 3 will verify the accuracy of the motion correction method using phantoms with simulated head motion. The goal is to ensure motion-corrected images meet performance metrics (e.g., spatial resolution, contrast-to-noise ratio) comparable to static images. By the end of Phase I, a robust software package capable of generating motion-corrected images for the portable, high-resolution TOF PET system within 5 minutes post- scan will be delivered. Phase II will refine the real-time motion correction software in imaging HNC patients and compare the diagnostic accuracy of the Head PETcoil to commercial whole-body PET systems. The Head PETcoil has the potential to transform HNC imaging by improving early-stage disease detection, enhancing treatment planning, and expanding access to advanced TOF PET imaging.