MONKEY: Inflammatory Cell Detection in Kidney Transplant Biopsies

Radboud University Medical Center provides 80 whole-slide images from four pathology institutes for the MONKEY challenge. The dataset contains Periodic acid-Schiff (PAS) stained images with annotated regions of interest for inflammatory cells, supporting AI model development for renal transplantation diagnostics. Up to three PAS scans and one IHC slide scan are available per whole-slide image.

Use Cases

• Train models for automated detection of monocytes and lymphocytes based on annotated regions of interest.
• Develop classification algorithms for inflammatory cells in PAS-stained whole-slide images.
• Explore multimodal image analysis using the available PAS and IHC slide scans.
• Benchmark AI models for reducing pathologists' workload in kidney biopsy assessment.

Strengths

• 80 whole-slide images provide a foundation for model training.
• Data originates from 4 different pathology institutes, suggesting potential diversity in sources.
• Includes multiple image modalities (PAS and IHC scans) per sample.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Radboud University Medical Center

Collection Method

Collected from 4 different pathology institutes, with annotated regions of interest.

Time Range

null

Freshness

null

Geography

null