Images of human tissue slides, some containing cancer, sourced from the NCT biobank and the UMM pathology archive. The dataset includes tissues from different parts of the body, organized into 9 classes. The dataset was created by Owkin and last updated on October 26, 2023.
Use Cases
- Classifying tissue types based on images from different body parts.
- Detecting cancerous tissue regions within slide images.
- Training computer vision models for medical image recognition.
- Benchmarking performance of AI models on a multi-class histopathology dataset.
Strengths
- Images are sourced from two authoritative medical institutions: the NCT biobank and the UMM pathology archive.
- The dataset is structured into 9 distinct tissue classes, which suggests a clear categorization.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- Last updated 2023-10-26 09:42:47; freshness should be verified.
Provenance
- Source
- NCT biobank (National Center for Tumor Diseases) and UMM pathology archive (University Medical Center Mannheim)
- Freshness
- 2023-10-26 09:42:47