Pneumonia Chest X-Ray Classification Challenge Dataset

Summary

• Introduction
• Dataset Structure
• Reference
• License

Introduction

This dataset is derived from the Labeled Optical Coherence Tomography (OCT) and Chest X-Ray Images for Classification collection on Mendeley Data, published by Daniel Kermany, Kang Zhang, Michael Goldbaum, and collaborators. It provides labeled chest radiographs for binary classification, containing images categorized as Normal or Pneumonia. The dataset is used in Pneumonia Chest X-Ray Classification Challenge, enabling participants to develop and evaluate machine learning models for medical image diagnosis.

Dataset Structure

Download

Download Pneumonia Chest X-Ray Data

Include

Participants will be provided with two datasets.

• Train set:

  • This dataset is used for training your machine learning model. We will provide .zip files containing Pneumonia chest X-ray data and a ground truth file with the following composition:
    • Normal (healthy lungs): 396 images.
    • Pneumonia (lungs with pneumonia): 918 images.
    • ground_truth.csv: Provides class labels corresponding to each image (label 0 is Normal and 1 is Pneumonia).

• Test set:

  • It contains .zip files of 1,171 unlabeled chest X-ray images for prediction submission.
  • Note:
    • The test data is used to predict labels for generating the submission file.
    • Submissions are evaluated and ranked on the Public Leaderboard.

Reference

For more dataset information, please go through the following link Pneumonia Chest X-Ray Data.

License

Creative Commons Attribution 4.0 International License - CC BY 4.0