A zero-dose synthetic baseline for the personalized analysis of [18F]FDG-PET: Application in Alzheimer’s disease

Abstract

PurposeBrain 2-Deoxy-2-[18F]fluoroglucose ([18F]FDG-PET) is widely used in the diagnostic workup of Alzheimer’s disease (AD). Current tools for uptake analysis rely on non-personalized templates, which poses a challenge as decreased glucose uptake could reflect neuronal dysfunction, or heterogeneous brain morphology associated with normal aging. Overcoming this, we propose a deep learning method for synthesizing a personalized [18F]FDG-PET baseline from the patient’s own MRI, and showcase its applicability in detecting AD pathology.MethodsWe included [18F]FDG-PET/MRI data from 123 patients of a local cohort and 600 patients from ADNI. A supervised, adversarial model with two connected Generative Adversarial Networks (GANs) was trained on cognitive normal (CN) patients with transfer-learning to generate full synthetic baseline volumes (sbPET) (192 × 192 × 192) which reflect healthy uptake conditioned on brain anatomy. Synthetic accuracy was measured by absolute relative %-difference (Abs%), relative %-difference (RD%), and peak signal-to-noise ratio (PSNR). Lastly, we deployed the sbPET images in a fully personalized method for localizing metabolic abnormalities.ResultsThe model achieved a spatially uniform Abs% of 9.4%, RD% of 0.5%, and a PSNR of 26.3 for CN subjects. The sbPET images conformed to the anatomical information dictated by the MRI and proved robust in presence of atrophy. The personalized abnormality method correctly mapped the pathology of AD subjects while showing little to no anomalies for CN subjects.ConclusionThis work demonstrated the feasibility of synthesizing fully personalized, healthy-appearing [18F]FDG-PET images. Using these, we showcased a promising application in diagnosing AD, and theorized the potential value of sbPET images in other neuroimaging routines.

Current Trends and Applications of PET/MRI Hybrid Imaging in Neurodegenerative Diseases and Normal Aging

“…Interestingly, there is an effort to generate zero-dose healthy baselines of FDG PET/MRI images and acquire hybrid imaging at ultra-low doses using GANs and CNNs to help the development of patient safety protocols [ 94 , 95 , 96 , 97 ]. Many clinicians interpret the FDG-PET component of hybrid images with healthy controls of a similar demographic.…”

“…The variational component allows the encoder to distribute the input onto a latent space allowing for variability and more control over how the VAE generates a custom image. In one study by Hinge et al, these artificially generated images were compared with healthy controls and had invariance to mean relative percent difference and mean absolute value relative difference [ 94 ]. Another set of projects focused on preserving diagnostic quality with shorter scan times with a high-sensitivity time-of-flight (TOF) PET system.…”

“…Amyloid and tau accumulation is the most common assessment performed in the evaluation of AD. A significant number of clinical studies within this review show an increased amyloid and tau protein burden in patients with different types of dementia [ 25 , 31 , 35 , 37 , 38 , 55 , 94 ]. These subtypes can be further differentiated depending on the local distribution of the amyloid.…”