Dear Sir or Madam,
I am not sure what training data is allowed for task 1. This years Pre-Treatment data and last years Post-Treatment data. But what about the additional_training_data which was also provided last year? Is it also allowed to use the additional_training_data for this challenge as training data?
Kind regards,
Bernhard Voggenberger
Created by Bernhard Miachel Voggenberger brats_participant Thank you very much for clarifying it for me!
Best regards,
Bernhard @brats_participant
That's true.
All the training data provided for [GLI-post](https://www.synapse.org/Synapse:syn59059776) and [GLI-pre](https://www.synapse.org/Synapse:syn65773245) can be used for model training.
Please note that the final testing phase will evaluate the Glioma segmentation collectively on both pre- and post-treatment data. You can find more details [here](https://www.synapse.org/Synapse:syn64153130/discussion/threadId=11986)
Best regards,
/Mehdi Thank you so much for your quick reply.
Just to make it as clear as possible for me. BraTS2024-BraTS-GLI-TrainingData.zip AND BraTS2024-BraTS-GLI-AdditionalTrainingData.zip are allowed for me to use for training?
Sorry for being so picky, I just want to be sure.
Kind regrards,
Bernhard Voggenberger Hi Bernhard,
Thanks for reaching out.
The original data of GLI 2025 includes both pre-treatment MRI (an updated dataset of BraTS GLI23) and post-treatment MRI (identical to BraTS GLI24).
As noted in the [Glioma Wiki page](https://www.synapse.org/Synapse:syn64153130/wiki/631053), you can use other datasets beyond the BraTS challenges:
_Participants may use publicly available annotated datasets, excluding those from other BraTS sub-challenges, and pre-trained models, excluding those generated using previous BraTS data. Use of private institutional data is allowed but cannot be part of the competition; it must be disclosed in the manuscript and accompanied by an additional MLCube during testing._
Please let us know if you have any questions,
/Mehdi