You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this model content.

Brant: Foundation Model for Intracranial Neural Signal

Brant is a foundation model in the field of intracranial recordings, which learns powerful representations of intracranial neural signals.

News

  • [2025.12.04] Considering that some users still have questions about the spectrum power calculation described in the paper, we have updated the code with the calculation process we designed for reference. Please refer to pretrain/pre_utils.

  • [2025.05.17] We have uploaded the model code on HuggingFace. Now anyone can access the finetuning code without sending email to the authors!

  • [2024.05.08] Due to the request from our collaboration partner, we are temporarily unable to directly share our code publicly. If you require access, please contact [email protected].

Source Code

The source code of Brant is provided in Brant_src.

  • Brant_src/train1.py, Brant_src/evaluate1.py are the code for the seizure detection task.
  • Brant_src/train2.py, Brant_src/evaluate2.py are the code for short- and long-term, frequency-phase forecasting tasks.
  • Brant_src/train3.py, Brant_src/evaluate3.py are the code for the imputation task.

Pre-trained weights

Disclaimer

The pre-training data for the Brant model was collected during routine treatment procedure of epilepsy patients from a hospital, and is intended solely for medical or research use. The pre-trained weights of Brant are released only for medical or research purposes and must not be subjected to any form of misuse.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support