Documents in Process

Magnetic resonance imaging (MRI) is highly susceptible to patient motion due to its relatively long acquisition times and the fact that data are acquired sequentially in k-space. Even small patient movements introduce phase inconsistencies across measurements, leading to severe artifacts such as blurring, ghosting, and geometric distortions that can compromise diagnostic quality. [...]

Plasma phosphorylated Tau217 (pTau217) represents a sensitive blood-based biomarker for Alzheimer’s disease (AD). This study investigated the performance of plasma pTau217 alone and as a composite score (Aβ1-40/Aβ1-42*pTau217) for detecting low Aβ1-42/1-40 in cerebrospinal fluid (CSF) as a surrogate marker of brain β-amyloid pathological changes.We analysed plasma samples from 82 pre-selected participants who were dichotomized according to their CSF Aβ42/40 ratio after data-driven cutoff determination by Gaussian mixture modelling. [...]

Group-equivariant convolutional neural networks (G-CNN) heavily rely on parameter sharing to increase CNN's data efficiency and performance. However, the parameter-sharing strategy greatly increases the computational burden for each added parameter, which hampers its application to deep neural network models. [...]