Positron emission tomography (PET) in vivo visualizes the molecular pathway and is the most sensitive molecular imaging modality routinely applied in clinic. Ionization radiation burden is a major concern in the practice of PET imaging, which hampers the application in many situation. Recent development in PET dramatically increased the effective sensitivity by increasing the geometric coverage leading to total-body PET imaging, which is confirmed to be able to reduce approximately 10 times radiation exposure. This encouraging breakthrough brings the hope of ultra-low dose PET imaging equivalent to transatlantic flight with the assistance of advanced computational methods. The proposed MICCAI Challenge aims to promote the development of computational methods, such as deep learning, to recover high-quality imaging from low-dose scans on total-body PET scanners. The reduction of radiation exposure towards as low as transatlantic flight will open new era for PET imaging and its clinical applications. This challenge will for the first time establish a benchmark dataset to tackle the technical challenges of ultra-lowdose imaging on total-body PET. It will integrate heterogeneous data acquired on two commercial total-body PET systems uExplorer (United Imaging) and Biograph Vision Quadra (Siemens Healthineers) to enhance the trustworthy of methodological developments and boost the clinical translational potential of the challenge outcome.