Official MIDL 2020 challenges

For the first time, MIDL invites the community to propose challenges involving both medical imaging and machine learning topics.


object-CXR

Automatic detection of foreign objects on chest X-rays

Analyzing chest X-rays is a common clinical approach for diagnosing pulmonary and heart diseases. However, foreign objects are occasionally presented on chest X-ray images, especially in rural and remote locations where standard filming guidelines are not strictly followed. Foreign objects on chest X-rays may obscure pathological finds, thus increasing false negative diagnosis. They may also confuse junior radiologists from real pathological findings, e.g. buttons are visually similar to nodules on chest X-ray, thus increasing false positive diagnosis. An algorithm based foreign object detection system could automatically promote re-filming, thus significantly reduce the cost and save radiologists time for more diagnosis.

We provide a large dataset of chest X-rays with strong annotations of foreign objects, and the competition for automatic detection of foreign objects. Specifically, 5000 frontal chest X-ray images with foreign objects (all manually annotated) as well as 5000 frontal chest X-ray images without foreign objects are provided.

More info: Challenge website


MC-MRRec

Multi-channel MR Image Reconstruction Challenge

Deep-learning-based Magnetic Resonance (MR) Imaging reconstruction has the potential to greatly speed-up MR exams. Our Multi-channel MR Image Reconstruction Challenge tackles this issue by:

  • Comparing different DL-based MR reconstruction models on a large dataset (> 200 GB)
  • Assessing reconstruction model generalizability to various datasets acquired with a different number of channels

The challenge has two separate tracks, and teams are free to decide whether to submit to just one track or both; we encourage teams to submit to both tracks. Each track will have a separate ranking.

In these two tracks, we will assess MR reconstruction quality and noticeable loss of high resolution detail, especially at the higher acceleration rates. By having two separate tracks, we hope to determine whether a generic reconstruction model applicable to various multi-channel coils will have a decreased performance (if any) compared to more coil-specific reconstruction models.

More info: Challenge website


More challenges to come.

The call for challenges is still available.