18+
Supported Datasets
Preprocessing, Training, and Downstream Analysis
NeuroSTORM provides end-to-end tools for fMRI preprocessing, self-supervised model training, and downstream analysis tasks across multiple public neuroimaging datasets.
5
Task Categories
Contrastive & MAE
Self-Supervised Pre-training
ROI & Volume
Analysis Modes
NeuroSTORM Capabilities
📊 Volume-Based Preprocessing
End-to-end preprocessing for volumetric fMRI data with skull stripping, registration, and brain extraction.
🎯 ROI-Based Analysis
Extract and analyze region-of-interest (ROI) time series with Harvard-Oxford atlases and custom parcellations.
🧠 Multiple Pre-training Methods
Support for SwiFT contrastive learning and MAE self-supervised methods on large-scale datasets.
🔧 Flexible Model Architecture
Customizable deep learning models with configurable heads for classification, regression, and embedding tasks.
📚 Comprehensive Datasets
Ready-to-use preprocessing pipelines and data loaders for UKB, HCP, ABCD, ADHD200, and more.
✅ Downstream Task Suite
Age/gender prediction, phenotype prediction, disease diagnosis, fMRI re-identification, and task state classification.
Explore More
Browse our datasets, tasks, and documentation to get started with NeuroSTORM.
NeuroSTORM Overview
