Connectome-based Hopfiled Neural Networks (fcHNN)
-
Updated
Mar 13, 2026 - Jupyter Notebook
Connectome-based Hopfiled Neural Networks (fcHNN)
Data from our own pre-processing of the ABIDE I sMRI dataset (1035 subjects) in FreeSurfer v6.
Early detection of Alzheimer’s Disease using 3D CNNs on structural MRIs, leveraging the ADNI dataset, Clinica preprocessing, and MeluXina HPC for high-performance training and Grad-CAM interpretability. 🧠
fMRI design optimization for complex experimental settings. Primary work: Atharv A. Umap and Valentin Guigon; Supervision: Valentin Guigon
Tutorial covering group DCM analyses of fMRI and M/EEG
Super-resolution of 3D electron microscopy images using deep learning, enabling high-quality reconstruction of large-scale 3D EM volumes across large brain fields of view.
"EEG-to-brain voxelization using hybrid information-theoretic methods (work in progress)"
Eigenvalue geometric features for cortical surface morphology analysis.
Advanced Multi-Omics Modeling for Elucidating Neuronal Architecture
This repository contains some codes for the computational neuroscience.
Fitting cable models (SWC) to 3D surface mesh and skeleton data
To associate your repository with the computational-neuroimaging topic, visit your repo's landing page and select "manage topics."