TC2-Res: A Structured Fusion of Tract-Level and Connectome-Level Brain Imaging in Athletes

A small-sample cohort of collegiate athletes provides brain imaging data combining diffusion tract descriptors with structural connectome descriptors. The dataset, authored by Yang Shi and last updated on 2026-06-03, accompanies a PDF describing the Tensor Connectome Consistency Residual (TC2-Res) fusion framework. This framework was evaluated on a primary football-vs.-others classification task.

Use Cases

• Benchmarking classical linear classifiers against learned multimodal fusion models based on tract and connectome descriptors.
• Developing anatomically structured fusion frameworks based on pair-aware tract representations and global modality branches.
• Investigating the alignment of matched anatomical families in a shared latent space using a consistency regularizer.
• Studying athlete-related brain imaging patterns in small cohorts using combined microstructural and organizational features.

Strengths

• Dataset is shared under a CC-BY-4.0 license, permitting broad reuse.
• Accompanies a published methodological framework (TC2-Res) for structured multimodal fusion.
• Focuses on a specific, challenging domain: small-sample neuroimaging studies of athletes.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The observed performance gain of the proposed TC2-Res framework was described as modest and inconsistent across folds.

Provenance

Source

figshare

Collection Method

Likely collected as part of a research study on collegiate athletes.

Freshness

Last updated 2026-06-03 05:42:29; freshness should be verified.