NeuMapper

NeuMapper leverages the Mapper algorithm from topological data analysis (TDA) to produce individualized, whole-brain representations of fMRI dynamics without requiring initial spatiotemporal averaging.

Key Features:

• Mapper / TDA-based representation: Applies the Mapper approach from topological data analysis to explore whole-brain configurations at the single-participant level.
• Dimensionality reduction elimination: Removes the need for dimensionality reduction commonly required in traditional Mapper implementations.
• Efficient parameter exploration: Implements a biologically grounded heuristic to facilitate efficient exploration of the Mapper parameter space.
• Integration with neuroanatomy and behavior: Uses novel meta-analytic approaches to anchor Mapper-derived representations to neuroanatomical structures and behavioral outcomes.
• Scalability and validation: Validated on multiple fMRI datasets including continuous multitask experiments and designed to scale to consortium-size datasets.

Scientific Applications:

• Single-participant fMRI analysis: Generate individualized insights into brain dynamical organization from fMRI data.
• Psychiatric neuroimaging and personalized medicine: Support research aimed at characterizing individual variability for personalized/precision approaches in psychiatry and neuroscience.
• Translational neuroimaging: Link neuroimaging-derived representations to behavioral outcomes to bridge imaging findings and clinical applications.
• Large-scale dataset analysis: Enable analysis of consortium-size fMRI datasets for population-level studies while preserving single-participant detail.

Methodology:

Uses the Mapper algorithm from topological data analysis; operates without initial spatiotemporal averaging or imposed dimensionality reduction; employs a biologically grounded heuristic for Mapper parameter exploration; applies meta-analytic anchoring of representations to neuroanatomy and behavior; validated on multiple fMRI datasets including continuous multitask experiments.