Brain Amyloid Burden Mapping Using MR Fingerprinting Aided by Deep Learning
Abstract
PURPOSE: To develop and externally validate a non-invasive framework for quantifying brain amyloid-β (Aβ) deposition using magnetic resonance fingerprinting (MRF) and neural network-based decoding, with positron emission tomography (PET) as the reference standard. METHODS: correction, covering the whole brain in 9.5 min. PET images were co-registered to the MRF space, and regional amyloid load was calculated using an automated template-based pipeline. An inverse mapping function was implemented to convert MRF signals into amyloid burden maps. Repeatability, agreement with PET-based centiloid values, and associations with cognitive scores were evaluated. RESULTS: The generated amyloid maps were visually similar to PET images. Test-retest analysis showed high repeatability, with a coefficient of variation of 1.8 ± 1.3% and an intraclass correlation coefficient of 0.84. In the external test set, MRF-based measurements correlated significantly with PET centiloid scores (Spearman's ρ = 0.589, P = 0.015) and Montreal Cognitive Assessment scores (ρ = -0.543, P = 0.020). CONCLUSION: The proposed framework enables non-invasive Aβ mapping using a clinically feasible MRI protocol and may support repeated assessment for monitoring during anti-amyloid treatment.
Related
- Peripheral DNA methylation of candidate Alzheimer’s disease risk genes and its associations with cognition, brain volume and amyloid-β burden
- Deep learning-based prediction of regional brain Tau PET burden from MRI.
- Contribution of Myelin Content to the PET-Measured Amyloid Burden
- Neurochemical fingerprinting of the Alzheimers disease continuum using 7 T short-echo-time brain single voxel 1H-MRS.
- Serial amyloid PET demonstrates marked reduction in amyloid burden following donanemab treatment: a real-world cohort study
BibTeX
@article{fujita2026brain,
title = {{Brain Amyloid Burden Mapping Using MR Fingerprinting Aided by Deep Learning}},
author = {Shohei Fujita and Yasutaka Fushimi and Yujiro Otsuka and Katsutoshi Murata and Guido Buonincontri and Gregor Koerzdoerfer and Mathias Nittka and Issei Fukunaga and Kaito Takabayashi and Akifumi Hagiwara and Yumiko Motoi and Madoka Nakajima and Koji Murakami and Atsushi Shima and Manabu Kubota and Berkin Bilgic and Koji Kamagata and Nobukatsu Sawamoto and Osamu Abe and Yuji Nakamoto and Shigeki Aoki},
journal = {Magnetic Resonance in Medical Sciences},
year = {2026},
month = aug,
doi = {10.2463/mrms.mp.2026-0112},
url = {https://www.jstage.jst.go.jp/article/mrms/25/4/25_mp.2026-0112/_article},
note = {CC BY},
}
RIS
TY - JOUR
TI - Brain Amyloid Burden Mapping Using MR Fingerprinting Aided by Deep Learning
AU - Shohei Fujita
AU - Yasutaka Fushimi
AU - Yujiro Otsuka
AU - Katsutoshi Murata
AU - Guido Buonincontri
AU - Gregor Koerzdoerfer
AU - Mathias Nittka
AU - Issei Fukunaga
AU - Kaito Takabayashi
AU - Akifumi Hagiwara
AU - Yumiko Motoi
AU - Madoka Nakajima
AU - Koji Murakami
AU - Atsushi Shima
AU - Manabu Kubota
AU - Berkin Bilgic
AU - Koji Kamagata
AU - Nobukatsu Sawamoto
AU - Osamu Abe
AU - Yuji Nakamoto
AU - Shigeki Aoki
PY - 2026
DA - 2026/08/21
JO - Magnetic Resonance in Medical Sciences
DO - 10.2463/mrms.mp.2026-0112
UR - https://www.jstage.jst.go.jp/article/mrms/25/4/25_mp.2026-0112/_article
AB - PURPOSE: To develop and externally validate a non-invasive framework for quantifying brain amyloid-β (Aβ) deposition using magnetic resonance fingerprinting (MRF) and neural network-based decoding, with positron emission tomography (PET) as the reference standard. METHODS: correction, covering the whole brain in 9.5 min. PET images were co-registered to the MRF space, and regional amyloid load was calculated using an automated template-based pipeline. An inverse mapping function was implemented to convert MRF signals into amyloid burden maps. Repeatability, agreement with PET-based centiloid values, and associations with cognitive scores were evaluated. RESULTS: The generated amyloid maps were visually similar to PET images. Test-retest analysis showed high repeatability, with a coefficient of variation of 1.8 ± 1.3% and an intraclass correlation coefficient of 0.84. In the external test set, MRF-based measurements correlated significantly with PET centiloid scores (Spearman's ρ = 0.589, P = 0.015) and Montreal Cognitive Assessment scores (ρ = -0.543, P = 0.020). CONCLUSION: The proposed framework enables non-invasive Aβ mapping using a clinically feasible MRI protocol and may support repeated assessment for monitoring during anti-amyloid treatment.
SN - 1880-2206
SN - 1347-3182
ER -