Alzheimer’s disease is the single most common thread running through CNS scholar projects. The work below approaches it from very different angles — electrophysiology, functional connectivity, graph learning, transfer learning — but shares a motivation: detecting change before clinical symptoms appear, using data already collected in routine practice.
CNS Projects¶
Diellor Basha — electrophysiological biomarkers of Abeta/tau burden in preclinical AD, combining MEG/EEG with MRI, PET and CSF measures from PREVENT-AD
Davor Curic — geometry-aware (tangent space) analysis of how the functional connectome moves over time, to track disease progression
Hao-Ting Wang — predicting presymptomatic features of AD with transfer learning, pooling large datasets to pretrain a model for a smaller clinical cohort
Josh Neudorf — graph neural network deep learning to find connectivity biomarkers of progression from mild cognitive impairment to dementia
Mohamed Abdelhack — shared and unique patterns of brain function between neurodegenerative and cardiovascular disease
Tom George — a neuro-foundation model pretrained on open data, fine-tuned for neurodegenerative disease prediction
Trishna Saha Detroja — sleep-related cortical cell-type subpopulations and their link to dementia-related brain change
Related Resources¶
Datasets — PREVENT-AD, ADNI, UK Biobank
Modalities — fMRI, EEG/MEG, PET, structural MRI
Methods — deep learning, graph neural networks, transfer learning
Related topics — Dementia, Aging and Cognitive Decline