A rigorous leave-one-subject-out benchmark on public auditory EEG data shows five-vowel decoding accuracy of 25.5 percent (chance 20 percent) using differential entropy features and LightGBM, with vowel information present but weak and localized to early auditory transients.
Diffusion tensor anisotropy in the cingulate gyrus in schizophrenia
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
representative citing papers
UniPET proposes a universal PET denoising network with style alignment network (SAN) and region-aware learning strategy (RALS) to handle varied dose reduction factors via domain generalization.
Unfolding Type-II Bayesian hyperparameter updates into a neural architecture with progressive correction mechanisms (biases to MLP to attention) improves reconstruction and convergence in brain source imaging while retaining algorithmic transparency.
Constructs hypergraphs from caCOH multivariate connectivity for EEG/MEG, recovering simulated coupling frequencies better than MSC graphs with reduction from 610 edges to 10 or 1 hyperedges.
A new NMF variant estimates integer and non-integer temporal shifts plus stretching in the frequency domain to improve brain tissue delineation in emission tomography data.
GNOVA reconstructs and forecasts CDR-SB and MMSE scores with MAEs of 1.35 and 2.28 on 1727 ADNI patients over 10 years using only routine visit data, enabling interpolation, extrapolation, and uncertainty estimates.
Empirical comparison of graded MRI preprocessing levels for MAE and JEPA pretraining on brain scans shows moderate levels (P2) are often sufficient, with limited additional utility from stronger preprocessing on downstream tasks.
PCA-derived global factor from gray-matter isotropic volume fraction in NODDI data correlates with specific cognitive scores in HCP Young Adult cohort.
fNIRS data and participant-independent ML models were used to estimate cognitive effort via relative neural efficiency and involvement, showing segment differences and close match between predicted and actual performance-derived effort.
Bootstrapped multi-level analysis of functional connectivity data reveals reduced DMN connectivity, increased deactivation, and 13 significantly differing regions (mostly frontal) in schizophrenia versus controls.
citing papers explorer
-
Reconstructing and forecasting disease trajectories of patients with Alzheimer's disease using routine data in resource-constrained settings
GNOVA reconstructs and forecasts CDR-SB and MMSE scores with MAEs of 1.35 and 2.28 on 1727 ADNI patients over 10 years using only routine visit data, enabling interpolation, extrapolation, and uncertainty estimates.