AI-driven PET reconstruction from histoimage

#Positron Emission Tomography (PET) generates functional images which represent biological processes (e.g., glucose metabolism, receptor affinity) occurring within a patient. According to PET techniques, a radiotracer is administered to a patient via intravenous injection, inhalation, oral ingestion or direct organ injection. Radioactive decay of the tracer as it travels within the patient generates positrons which eventually encounter electrons and are annihilated thereby.

#An annihilation produces two 511 keV photons which travel in approximately opposite directions. A coincidence is identified when two detector crystals detect the arrival of two photons within a short time window indicating that the two photons arose from the same positron annihilation. Because the two “coincident” photons travel in approximately opposite directions, the locations of the two detector crystals determine a Line-of-Response (LoR) along which an annihilation may have occurred. PET data represents each detected annihilation as a LoR between two detector crystals.

#An image is reconstructed from the PET data using known algorithms such as filtered backprojection (FBP) and ordered subsets expectation maximization (OSEM). PET image reconstruction is complex, resource-intensive and time-consuming. For example, reconstruction usually includes determination and subtraction of random coincidences and scatter coincidences from the PET data, attenuation correction, motion correction, correction for system sensitivity, and other suitable steps.

#FIG. 1 is a block diagram of a system to generate a corrected PET image according to some embodiments.

#FIG. 2 is a flow diagram of a process to generate a corrected PET image according to some embodiments.

#FIG. 3 illustrates training of a neural network to generate a scatter histoimage based on a normalized histoimage according to some embodiments.

#The following description is provided to enable any person in the art to make and use the described embodiments. Various modifications, however, will remain apparent to those in the art.

#Embodiments utilize a trained neural network to derive a PET image from a normalized PET histoimage and a corresponding anatomical image. Embodiments may thereby generate fully-corrected, low-noise, high spatial resolution PET images in a practical timeframe using typically-available computing resources.

#A PET histoimage is a simple backprojection of each detected annihilation with Time-of-Flight (ToF) localization along an LoR. Briefly, list-mode data is acquired and normalized based on scanner sensitivity. The list-mode data may also be attenuation-corrected based on previously-acquired anatomical information. A normalized PET histoimage is generated from the normalized list-mode data. A scatter histoimage is generated based on the PET histoimage and a corresponding mu-map, and the scatter histoimage is used to scatter-correct the PET histoimage. The scatter-corrected PET histoimage and an anatomical image (e.g., a Computed Tomography (CT) image) are then input to a trained neural network to generate a corrected PET image.