A Multi-Stain Breast Cancer Histological Whole-Slide-Image Data Set from Routine Diagnostics

Area of Science:

• Pathology
• Computational Biology
• Oncology

Background:

• Hematoxylin and eosin (H&E) and immunohistochemistry (IHC) staining of FFPE breast cancer tissue sections are crucial for pathological assessment.
• IHC is widely used for biomarker status (ER, PGR, HER2, KI67) in diagnostics.
• Computational pathology shows promise for biomarker assessment using whole slide images (WSIs).

Purpose of the Study:

• To address the scarcity of public data for computational pathology in breast cancer research.
• To create the largest publicly available dataset of matched H&E and IHC WSIs from primary breast cancer specimens.

Main Methods:

• Collected FFPE tissue sections from surgical resections of female primary breast cancer patients.
• Stained sections with H&E and various IHC markers.
• Acquired whole slide images (WSIs) for all stained sections.
• Matched WSIs from H&E and IHC stains from the same tumor tissue.

Main Results:

• Published the largest public dataset to date containing matched H&E and IHC WSIs.
• The dataset comprises 4,212 WSIs from 1,153 primary breast cancer patients.
• This dataset represents a significant resource for developing and validating computational pathology algorithms.

Conclusions:

• The newly released dataset significantly enhances the availability of high-quality, matched imaging data for breast cancer research.
• Facilitates the development of advanced computational tools for more accurate and efficient biomarker analysis.
• Aims to accelerate progress in precision medicine for breast cancer through improved image analysis techniques.