Classification of amyloid-β status using 3D quantitative-amplified Magnetic Resonance Imaging (3D q-aMRI) - A preliminary study

Motivation: 3D q-aMRI is a technique for visualizing and quantifying pulsatile sub-voxel displacement in the brain. This study investigates the potential of pulsatile brain motion as a predictor of amyloid-β (Aβ) status. Goal(s): Assess the potential of 3D q-aMRI in differentiating between Aβ-positive and Aβ-negative patients Approach: The sub-voxel displacement field output from 3D q-aMRI was utilized in 46 patients -both Aβ+ and Aβ- -to train a multi-layer perceptron for classification purposes. Results: Classification accuracy reached 74.2%, despite the limited sample size and use of a basic multi-layer perceptron (MLP) model, indicating detectable differences in brain motion between Aβ+ and Aβ- cases. Impact: Abnormal amyloid-β (Aβ) buildup signals early Alzheimer's but requires invasive assessments. This study used 3D q-aMRI to classify Aβ status via pulsatile brain motion, achieving 74.2% accuracy with a basic MLP model trained on sub-voxel displacement data.