Graph-NuCLS: Cell-Graph Dataset for Nucleus Classification

Graph-NuCLS is a node-level classification dataset derived from the NuCLS dataset. Each tissue patch is converted into a cell-graph where nodes represent detected cell nuclei and edges encode spatial proximity. The dataset was authored by ogutsevda and last updated on 2026-03-03.

Use Cases

• Train node classification models based on cell-graph structure.
• Benchmark graph machine learning algorithms for histopathology tasks.
• Study relationships between cell morphology, texture, and spatial arrangement.
• Develop models for automated nucleus type prediction in tissue samples.

Strengths

• Derived from the established NuCLS dataset, suggesting a foundation in real medical data.
• Node features describe cell morphology, texture, and color intensity, providing multi-modal signals.
• Task involves predicting across 7 distinct cell type classes.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Freshness should be verified as the last update timestamp is in the future (2026-03-03).

Provenance

Source

Derived from the NuCLS dataset.

Collection Method

Tissue patches converted into cell-graphs with nodes as nuclei and edges for spatial proximity.

Freshness

Last updated 2026-03-03 09:00:09