MCI-specific latent features

1. MCI-specific latent features predict Alzheimer's disease in silico (AUC = 0.83).

   “Moreover, predictive modeling using these features achieved superior AD-conversion prediction (AUC = 0.83) versus whole-brain atrophy (AUC = 0.74; p < 0.001) or CSF biomarkers (AUC = 0.77; p < 0.001).”

   In silicobrain

   Decoupling MCI-specific signatures from shared neurobiological substrates of cognitive aging via deep learning

   NPJ digital medicine·23 Apr 2026

More on Alzheimer's disease