deepbci/deep-bci

An open software package to develop BCI based brain and cognitive computing technology for recognizing user's intention using deep learning

Deep BCI SW ver. 1.0 is released.

An open software package to develop Brain-Computer Interface (BCI) based brain and cognitive computing technology for recognizing user's intention using deep learning

Website: http://deepbci.korea.ac.kr/

We provide detailed information in each forder and every function.

1. 'Intelligent_BCI': contains deep learning-based intelligent brain-computer interface-related function that enables high-performance intent recognition.

• Domain Adversarial NN for BCI: functions related to domain adversarial neural networks
• EEG based Meta RL Classifier: functions related to model-based reinforcement learning
• GRU based Large Size EEG Classifier: data and functions related to gated recurrent unit
• etc

2. 'Ambulatory_BCI': contains general brain-computer interface-related functions that enable high-performance intent recognition in ambulatory environment

• Channel Selection Method based on Relevance Score: functions related to electrode selection method by evaluating electrode's contribution to motor imagery based on relevance score and CNNs
• Correlation optimized using rotation matrix: functions related to cognitive imagery analysis using correlation feature
• SSVEP decoding in ambulatory environment using CNN: functions related to decoding scalp- and ear-EEG in ambulatory environment
• etc

3. 'Cognitive_BCI': contains cognitive state-related function that enable to estimate the cognitive states from multi-modality and user-custermized BCI

• multi-threshold graph metrics using a range of critiera: functions related to entrain brainwaves based on a combined auditory stimulus with a binaural beat
• EEG_Authentication_Program: identifying individuals based on resting-state EEG
• Ear_EEG_Drowsiness_Detection: identifying individuals based on resting-state EEG using convolutional neural network
• etc

4. 'Zero-Training_BCI': contains zero-training brain-computer interface-related functions that enable to minimize additional training

• ERP-based_BCI_Algorithm_for_Zero_Training: functions related to Event Related Potential (ERP) analysis including feature extraction, classification, and visualization
• SSVEP_based_Mind_Mole_Catching: functions allowing users to play mole cathcing game using their brain activity on single/two-player mode
• SSVEP_based_BCI_speller: functions related to SSVEP-based speller containing nine classes
• etc

Acknowledgement: This project was supported by Institute for Information & Communications Technology Promotion (IITP) grant funded by the Korea government(No. 2017-0-00451, Development of BCI based Brain and Cognitive Computing Technology for Recognizing User’s Intentions using Deep Learning).

deep-bci's Issues

question: datapath and filename

Hi,
I'm checking on SSVEP code and confirming that the 'datapath' and 'filename' in SSVEP_CCA_Classification and SSVEP_FFT_Visualization are different.
I think this part should be revised.

Thank you.

Inter_Subject_Contrastive_Learning_for_EEG------Nan Loss

Features become all Nan after self.rnn layer, and loss is nan.

which you're device or hardware information ?

RT

please

Where can I find the funcition named eegfile_loadBV?

Where can I find a function called eegfile_loadBV in the file that exists in the link below?

https://github.com/DeepBCI/Deep-BCI/tree/master/2_1_Ambulatory_BCI
/Motor_imagery_on_treadmill/

No code about Inter-subject Contrastive Learning for SubjectAdaptive EEG-based Visual Recognition

Why have I not seen the code of this paper: Inter-subject Contrastive Learning for SubjectAdaptive EEG-based Visual Recognition?If it is convenient, could you please upload it again?

Bcu