Pancreatic Tissue Histology Images: 90 Normal vs. Glucose-Intolerant Samples

90 RGB microscopy images of pancreatic tissue acquired at 100x magnification with a resolution of 1280x960 pixels. The dataset includes 58 images of normal pancreatic tissue and 32 images of glucose-intolerant tissue, intended for research in medical image analysis and supervised learning. It was contributed by authors L. Rato, F.C. Silva, A.R. Costa, and C. Antunes under a CC-BY-4.0 license.

Use Cases

• Binary classification of pancreatic tissue based on histology images.
• Training supervised learning models for medical image diagnosis.
• Researching explainable AI methods for microscopy image analysis.
• Benchmarking low-data classification algorithms in a medical context.

Strengths

• Contains 90 labeled images, providing a defined scope for binary classification tasks.
• Includes 58 normal and 32 glucose-intolerant samples, establishing a clear class distribution.
• Images have a consistent resolution of 1280x960 pixels and were acquired at 100x magnification.

Limitations

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

Provenance

Source

openml

Collection Method

Microscopy images of pancreatic tissue acquired at 100x magnification.

Time Range

null

Freshness

Last updated date is unknown.

Geography

null