PCPAm - A dataset of histopathological images of penile cancer for classification tasks

This dataset, developed by the Cancer Project of the Amazon Region with contributions from faculty and students of the Applied Computing Group at the Federal University of Maranhão (UFMA), comprises 194 high-resolution RGB histopathological images (2048x1536 pixels). This dataset is specifically categorized by magnification levels (40× and 100×) and pathological classification (Tumor or Non-Tumor).

Summary of Image Distribution

• Pathological Classification: Tumor images represent squamous cell carcinoma, with 55 images at each magnification level. Non-Tumor images are evenly distributed, with 42 images at both magnification levels.
• HPV Status: The dataset includes 15 images labeled as HPV-positive and 30 as HPV-negative, equally distributed across both magnification levels.
• Histological Grades: The dataset comprises 5 images classified as Grade 1, 25 images as Grade 2, and 20 images as Grade 3, illustrating the diversity of cancer progression stages across both magnification levels.

Dataset Contents

The dataset is organized into the following structure:

• images/: A folder containing all histopathological images in JPEG format.
• data_information.xlsx: A comprehensive file detailing each image, including:
  • Slide
  • File Name
  • Magnification Level (40× or 100×)
  • Pathological Classification (Tumor or Non-Tumor)
  • HPV Status (Positive or Negative)
  • Histological Grade (Grade 1, Grade 2, or Grade 3)

Notice of Use

This dataset is available under a Creative Commons CC BY-NC-ND license (Attribution-NonCommercial-NoDerivs).
You can use this dataset for research purposes, provided that proper attribution is given to our work and publications. However, modification of the dataset or its use for commercial purposes is strictly prohibited.