TCGA-BRCA Breast Cancer Slide Images with Malignant and Non-Malignant Annotations

Zahra Alidousti Shahraki released a dataset of annotated breast cancer whole slide images from the TCGA-BRCA public dataset. The 61.6 GB collection includes pathologist-annotated regions, patches extracted from those regions, and a separate set of tumor and non-tumor patches. It was last updated on 2026-05-27.

Use Cases

• Train image classification models to distinguish malignant from non-malignant tissue based on annotated slide regions.
• Develop patch-based analysis algorithms using the extracted patches from annotated whole slide images.
• Benchmark segmentation models against pathologist-drawn annotations for tumor regions.
• Create synthetic data or augment training sets using the provided tumor and non-tumor patch collections.

Strengths

• Includes pathologist-annotated regions, providing a ground truth for malignant vs. non-malignant tissue.
• Contains a large volume of data at 61.6 GB, likely offering substantial training material.
• Derived from the established TCGA-BRCA public dataset, suggesting a reputable underlying source.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for specific modeling tasks.
• Description metadata is limited; actual data quality and annotation consistency require manual inspection after download.

Provenance

Source

TCGA-BRCA public dataset, annotations by Zahra Alidousti Shahraki.

Collection Method

Pathologist annotations on whole slide images, followed by patch extraction.

Freshness

Last updated 2026-05-27 07:01:26; freshness should be verified.