CGM-JEPA Downstream: Labeled Cohort Splits for Metabolic Outcome Evaluation

Labeled cohort splits used to evaluate CGM encoders on two binary metabolic outcomes — insulin resistance and β-cell dysfunction. The dataset was created by CRUISEResearchGroup for the paper CGM-JEPA: Learning Consistent Continuous Glucose Monitor Representations via Predictive Self-Supervised Pretraining. It was last updated on HuggingFace on 2026-05-13.

Use Cases

• Evaluate encoder performance on predicting insulin resistance based on labeled cohort data.
• Benchmark models for β-cell dysfunction classification using the provided evaluation splits.
• Assess the generalizability of pretrained CGM representations on downstream clinical tasks.

Strengths

• Dataset is specifically designed for evaluating encoders on two defined binary metabolic outcomes.
• Linked to a published research paper and associated pretraining corpus, providing context.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

CRUISEResearchGroup

Collection Method

Created as downstream evaluation splits for a research paper.

Freshness

Last updated 2026-05-13 11:30:52; freshness should be verified.