Dear Team, I have a question about the BRATS competition. We are considering integrating a pre-trained VGG model into our architecture solely for the purpose of loss calculation. Is this approach permissible? Thank you for your guidance. Best regards, Sarim

Created by Sarim Hashmi ghost123
Can you provide more context? For which subchallenge and pretrained on what exactly? See Below rules regarding Use of Data Beyond BraTS Use of Data Beyond BraTS: Participants are NOT allowed to use additional public and/or private data (from their own institutions) for extending the provided data, for the training of the algorithm chosen to be ranked. Similarly, using models that were pretrained on such datasets is NOT allowed. This is due to our intentions to provide a fair comparison among the participating methods. However, participants are allowed to use additional public and/or private data (from their own institutions), only if for scientific publication purposes and they explicitly mention this in their submitted manuscripts and also report results using only the Challenge data to discuss potential result differences.

Use of pretrained model to calculate loss in the model architecture page is loading…