Amyloid PET-Like Image Generation Technology
MR Fingerprinting × AI
Plusman, in collaboration with Juntendo University, has developed an amyloid PET-like image generation technology.
A presentation on this technology by Dr. Shohei Fujita was awarded 1st Place Winner in the Clinical Translation: Unmet Needs Challenge at the International Society for Magnetic Resonance in Medicine (ISMRM) Annual Meeting & Exhibition 2023.
Shohei Fujita, et al., MR fingerprinting for quantification of brain amyloid burden: from development to prospective multi-site external validation, ISMRM23, 2023
Research Overview
Introduction
Amyloid beta (Aβ)-PET is important for objective detection of Aβ deposition and definitive diagnosis of Alzheimer’s disease . However, limited access, high cost, undesirable injections, and radiation exposure restrict the applicability of Aβ-PET scans. As new amyloid-reducing drugs are being developed, there is a growing need for non-invasive Aβ imaging techniques. In this study, we aimed to develop a non-invasive Aβ imaging technology and evaluate its clinical utility for the objective diagnosis and treatment monitoring of Alzheimer’s disease.
Methods and Results
In this study, we developed a framework that utilizes efficient MR fingerprinting (MRF) data acquisition and complex-valued neural network based decoding to quantify Aβ in the brain. Using this framework, we analyzed data from both Aβ-positive and Aβ-negative patients, obtaining outputs (MRF amyloid maps) that were visually similar to Aβ-PET images (Figure 1). External validation of the test set compared MRF-based scoring with amyloid PET and cognitive function tests, and significant correlations were found between MRF SUVr and the Aβ PET centiloid score (Spearman rho correlation coefficient = 0.589, p=.015), as well as the MoCA score (-0.543, p=.020). Additionally, the method discriminated subject-level Aβ positivity with an AUC of 0.84 (Figure 2).
