Description
For this challenge we provide you with the data and give you complete freedom in what you choose to do and how you choose to predict whether that image has pneumonia or not. The field of healthcare is constantly growing and we want to give our community a feel of the environment. In this challenge you must create a model that will read an image (black and white) and output whether or not that image has pneumonia or not. Keep in mind that this image is a frontal x-ray of the lungs. These images are also in nih format, so the processing of this data is a tad more difficult. Feel free to discuss any questions you have on the dicussion forum and group up with people to do this competition.
Input
Your binary classification model will be given a cxr black and white image as input. This cxr is a frontal xray of the lungs with may or may not contain the disease pneumonia. The cxr images real value along with other additional information is stored in a csv file. In the data section below you can download and examine the data. (Source: Kaggle)
Output
Your binary classification model should return a boolean value, True (1) or False (0). True (1) means that there is pneumonia visible in the CXR. False (0) means that there is no pneumonia in the CXR.
Examples
Please follow these instructions to download and access the data.
Scroll down to the bottom of the page
Click the download all button
Once you have downloaded the data, to save space you can continue to do these optional steps
Delete the following files/folders as those are not needed for this challenge (stage_2_test_images, stage_2_sample_submission.csv, stage_2_detailed_class_info.csv, GCP Credits Request Link - RSNA.txt)