CLAHE Enhanced Mammography Images from CBIS-DDSM

CBIS-DDSM Mammography is a machine-learning-ready version of the Digital Database for Screening Mammography. It has been heavily engineered and preprocessed, specifically enhanced using Contrast Limited Adaptive Histogram Equalization (CLAHE). The original database is a widely-used benchmark for breast cancer screening research.

Use Cases

• Train a convolutional neural network to classify mammogram images as benign or malignant.
• Develop an object detection model to localize masses and calcifications within the enhanced mammography images.
• Benchmark image segmentation algorithms on the preprocessed, CLAHE-enhanced mammogram regions.
• Evaluate the impact of CLAHE preprocessing on model performance for breast lesion classification.

Strengths

• Based on the widely-used CBIS-DDSM benchmark dataset for mammography analysis.
• Images are preprocessed and enhanced with CLAHE, making them ready for machine learning pipelines.

Limitations

• Specific row counts, image dimensions, and class distribution details are unavailable.
• The dataset is image-only, lacking associated patient metadata or structured tabular data.

Provenance

Source

Curated from the Curated Breast Imaging Subset of the Digital Database for Screening Mammography (CBIS-DDSM).

Collection Method

Heavily engineered and preprocessed version of the original database, with CLAHE enhancement applied.