ML-CDS 2020: Challenge

The challenge is related to the detection and recognition of tubes and lines in AP Chest X-rays.

Chest X-rays are the most common diagnostic exams in emergency rooms and hospitals. There has been a surge of work on automatic interpretation of chest X-rays using deep learning approaches after the availability of large open source chest X- ray dataset from NIH. However, the labels are not sufficiently rich and descriptive for training classification tools.

Further, it does not adequately address the findings seen in Chest X- rays taken in anterior-posterior (AP) view which also depict the placement of devices such as central vascular lines and tubes. In this challenge, we have relabeled a portion of the NIH dataset for the presence and recognition of tubes and lines.

The challenge consists of two tasks, namely:

  1. Detection of the presence of lines and tubes
  2. Recognition of the line/tube type when present in chest X-rays

All the labels provided in this challenge were generated by our team for a subset of images from the public CheXNet NIH dataset. In the links below, only the labels are provided, where the images themselves need to be downloaded from the NIH website in the following link: Here.

As this dataset was initially published in a paper at ISBI'19 please be sure to cite this reference in your papers on this challenge data -

The deadlines for the challenge are as follows:

  1. Registration opens: May 4, 2020
  2. Challenge models submission deadline: June 26, 2020
  3. Challenge paper submission deadline: July 1, 2020
  4. Challenge winner notification deadline: August 1, 2020
  5. Challenge presentation: October 4, 2020

Note: The models can be updated until the submission deadline.