-
Notifications
You must be signed in to change notification settings - Fork 1.1k
Evaluation, Reproducibility, Benchmarks Meeting 16
AReinke edited this page May 30, 2022
·
2 revisions
Date: 25th May 2022
Present: Carole, Michela, Jorge, Keyvan, Lena, Annika
- Metrics Reloaded initiative - Metrics implementation:
- Annika and Carole will check which of the metrics are already implemented in MONAI and if we need some update
- Annika will send the prefinal metric table to Carole
- Carole will provide the metric table and results of the analysis to the Dev team
- Contact person for multi-threshold and counting metrics: Paul Jäger
- Contact person for distance-based metrics: Carole
- Within MONAI, the metrics should be grouped by metric families
- MONAI Bundle (Configuration of models, including metadata, pt files, ..)
- Reproducibility is desired at different levels
- Sharing weights/pretrained models
- Sharing models that are ready to be used
- Sharing models that are validated for a clinical application that are ready to be used
- We may have a simple Delphi process after summer
- We will discuss this topic in the next MICCAI SIG Board meeting (June 20th)
- Reproducibility is desired at different levels
- MONAI datasets
- The main work should be done by MONAI’s data working group
- Once they converged to a solution, we will make the connection to challenge organizers
- Dan and Jens will leave the WG