MICCAI MS and PET Segmentation Challenges

Challenges description

The aim of the “MICCAI 2016” challenges organized by FLI-IAM is to demonstrate the relevance of our # 2 scenario in real life and to exhibit how IT e-infrastructure is able to control the evaluation of segmentation algorithm on medical images. These challenges provide assess the state of the art segmentation methods with tools (services) provided by participants and running on data and infrastructure provided by the FLI-IAM node. The databases were provided by communities of users together with FLI-IAM : OFSEP for the competition in Multiple Sclerosis and the American Association of Physicists in Medicine (AAPM and its working group 211) for the  PET competition. The image processing solutions are uploaded by the participant to the FLI-IAM infrastructure and executed automatically to the data. The results are automatically recorded on the data server of FLI-IAM  infrastructure in addition with the  evaluation metrics. In summary, the FLI e-Infra is used for::

  • Host the imaging data of the challenges
  • Host and run the image processing tools of the challenges
  • Calculate the results of the challenges
  • Hosts the data processed by the different methods
  • Disseminate the results of the challenges

A web portal was implemented at that time to manage different aspects (of user authentication navigation in infrastructure services).

Links:

[1] OFSEP : Observatoire Français pour la Sclérose en Plaques (http://www.ofsep.org)

[2] ]http://aapm.org/org/structure/default.asp?committee_code=TG211

Comments are closed.