After uploading the Docker file for Preliminary Test Task 2, I encountered a submission failure. However, following the guide provided in this post: https://grand-challenge.org/forums/forum/synthrad2023-643/topic/debugging-algorithm-docker-submissions-after-submitting-on-the-challenge-1478/, I couldn't find any cases listed in the Results tab of my algorithm, so I'm unable to view any log information. I would like to know how to resolve this issue.

In addition, I would like to clarify a few points:

  1. I noticed that the task description mentions that it's possible to use different models for different organs. However, the submission interface does not seem to distinguish between organs. Does this mean I need to determine the organ type from each case by reading the filename and then select the corresponding model accordingly?

  2. I did not integrate my algorithm into the provided template. Instead, I created the Docker container based on my own framework (the test.sh script from the template runs successfully with it). My algorithm does not process the 3D input image directly. Instead, it slices the image into 2D slices, processes them through the model, saves the predictions locally, then reconstructs the output as a 3D image, and deletes the intermediate 2D results. The final output saved in /output/ is correct, although the workflow differs from the example provided in the template.

  3. Does the content of results.json in the /output/ directory affect whether the submission is considered successful? Or is it only checked whether the generated CT exists in the synthetic-ct/ folder? I'm referring here to the actual evaluation logic after submitting the Docker container on the website, not the behavior in the test.sh script in the template.