MIAS Mammography Dataset for Breast Cancer Detection

The MIAS Mammography Dataset is a comprehensive collection of mammogram images designed to aid research in breast cancer detection. The dataset consists of labeled images that include various types of abnormalities such as calcifications, spiculated masses, and architectural distortions, making it ideal for training machine learning models and AI systems in medical image analysis.

Key Features of the Dataset:

• MIAS No (Reference Number): Unique identifier for each mammogram image in the database.
• Background Tissue (BG): The type of background tissue in the mammogram images, classified as:
  • F: Fatty
  • G: Fatty-Glandular
  • D: Dense-Glandular
• Abnormality Class (CLASS): Type of abnormality identified in the mammogram, including:
  • CALC: Calcification
  • CIRC: Well-Defined Masses
  • SPIC: Spiculated Masses
  • MISC: Other, Ill-Defined Masses
  • ARCH: Architectural Distortion
  • ASYM: Asymmetry
  • NORM: Normal
• Severity (SEVERITY): The severity of the abnormality, marked as:
  • B: Benign
  • M: Malignant
• Coordinates & Radius: The (x, y) coordinates marking the center of the abnormality and the radius of the enclosing circle.
• Density (DENSITY): Tissue density classification from A (low) to D (high).
• BI-RADS Classification: Used for assessing the severity of abnormalities (e.g., BI-RADS 1 for normal, BI-RADS 5 for highly suspicious malignancy).
• Group Classification: General categories such as Normal, Masses, or Calcification.

Usage and Licensing: The MIAS Mammography Dataset is provided by the Mammographic Image Analysis Society (MIAS) for research purposes only. It is protected under copyright laws and can be used to develop AI models, conduct studies, and enhance breast cancer detection tools. The dataset cannot be rented, leased, or sold, and any publications resulting from its use must reference the dataset.