Federated Transfer Learning For EEG Signal Classification

Authors: Ce Ju, Dashan Gao, Ravikiran Mane, Ben Tan, Yang Liu and Cuntai Guan

Published in: 42nd Annual International Conferences of the IEEE Engineering in Medicine and Biology Society in conjunction with the 43rd Annual Conference of the Canadian Medical and Biological Engineering Society (EMBS) July 20-24, 2020 via the EMBS Virtual Academy

---

Acknowledgements

Huge thanks to great geometers including Bernhard Riemann, Shiing-Shen Chern, Shing-Tung Yau, etc. Because of their noble contributions, we see the world differently.

Introduction

The success of deep learning (DL) methods in the Brain-Computer Interfaces (BCI) field for classification of electroencephalographic (EEG) recordings have been restricted by the lack of large datasets. Privacy concerns associated with the EEG signals limit the possibility to construct a large EEG- BCI datasets by the conglomeration of multiple small ones for jointly training the machine learning model. Hence, in this paper, we propose a novel privacy-preserving DL architecture named federated transfer learning (FTL) for EEG classification that is based on the federated learning framework. Working with the single-trial covariance matrix, the proposed architecture extracts common discriminative information from multi-subject EEG data with the help of domain adaption techniques. We evaluate the performance of the proposed architecture on the PhysioNet dataset for 2-class motor imagery classification. While avoiding the actual data sharing, our FTL approach achieves 2% higher classification accuracy in a subject-adaptive analysis. Also, in the absence of multi-subject data, our architecture provides 6% better accuracy compared to other state-of-the-art DL architectures.

---

Network Architecture

Our proposed architecture consists of following 4 layers: manifold reduction layer, common embedded space, tangent projection layer and federated layer. The purpose of each layer is as follows:

1. Manifold reduction layer: Spatial covariance matrices are always assumed to be on the high-dimensional SPD manifolds. This layer is the linear map from the high-dimensional SPD manifold to the low-dimensional one with undetermined weights for learning.

2. Common embedded space: The common space is the low-dimensional SPD manifold whose elements are reduced from each high-dimensional SPD manifolds, which is designed only for the transfer learning setting.

3. Tangent projection layer: This layer is to project the matrices on SPD manifolds to its tangent space, which is a local linear approximation of the curved space.

4. Federated layer: Deep neural networks are implemented in this layer. For the transfer learning setting, parameters of neural networks are updated by the federated aggregation.

---

How To Run The Code

Please install pyRiemann and mne

pyRiemann: https://github.com/alexandrebarachant/pyRiemann mne: https://github.com/mne-tools/mne-python

The SPD-Net part is originally from https://github.com/YirongMao/SPDNet

Data pre-processing

Please put your training data and labels into a directory "raw_data/" in this project. The package mne is adopted for EEG data pro-processing. To generate the required data as SPDNet input, please refer to the following example code:

from mne import Epochs, pick_types, events_from_annotations
from mne.io import concatenate_raws
from mne.io.edf import read_raw_edf
from mne.datasets import eegbci

# Set parameters and read data

# avoid classification of evoked responses by using epochs that start 1s after
# cue onset.
tmin, tmax = 1., 2.
event_id = dict(hands=2, feet=3)
subject = 7
runs = [6, 10, 14]  # motor imagery: hands vs feet

raw_files = [
    read_raw_edf(f, preload=True) for f in eegbci.load_data(subject, runs)
]
raw = concatenate_raws(raw_files)

picks = pick_types(
    raw.info, meg=False, eeg=True, stim=False, eog=False, exclude='bads')
# subsample elecs
picks = picks[::2]

# Apply band-pass filter
raw.filter(7., 35., method='iir', picks=picks)

events, _ = events_from_annotations(raw, event_id=dict(T1=2, T2=3))

# Read epochs (train will be done only between 1 and 2s)
# Testing will be done with a running classifier
epochs = Epochs(
    raw,
    events,
    event_id,
    tmin,
    tmax,
    proj=True,
    picks=picks,
    baseline=None,
    preload=True,
    verbose=False)
labels = epochs.events[:, -1] - 2

# cross validation
cv = KFold(n_splits=10, random_state=42)
# get epochs
epochs_data_train = 1e6 * epochs.get_data()

# compute covariance matrices
cov_data_train = Covariances().transform(epochs_data_train)

Model training

For subject-adaptive analysis, run SPDNet_Federated_Transfer_Learning.py

For subject-specific analysis, run SPDNet_Local_Learning.py

---

Federated Transfer Learning Framework for Biomedical Applications

We developed a federated transfer learning framework for various biomedical applications. The FTL framework adopts several federated learning frameworks such as FATE and pyTorch. It supports biomedical machine learning tasks with different types of data. We believe the FTL framework provides an easy-to-learn tool for researchers to study machine learning tasks of different biomedical data with privacy protection and good performance. The FTL framework will be open-sourced soon.

---

Contributions

We apply federated learning framework to build BCI models from multiple subjects with heterogeneous distributions.

1. FTL preserves user privacy by keeping EEG data of each subject on-device.
2. FTL achieves 6% better accuracy than other state-of-the-art DL methods in EEG-MI task, by using spatial covariance matrix as an input.
3. FTL achieves the best classification accuracy over bad subjects via transfer learning.

---

Cite Us

For detailed information about our work, please refer to our paper published in 42nd Annual International Conferences of the IEEE Engineering in Medicine and Biology Society in conjunction with the 43rd Annual Conference of the Canadian Medical and Biological Engineering Society (EMBS) July 20-24, 2020 via the EMBS Virtual Academy:

Federated Transfer Learning for EEG Signal Classification

If this project helps you in your research, please cite our work in your paper.

@article{ju2020federated,
  title={Federated Transfer Learning for EEG Signal Classification},
  author={Ju, Ce and Gao, Dashan and Mane, Ravikiran and Tan, Ben and Liu, Yang and Guan, Cuntai},
  journal={IEEE Engineering in Medicine and Biology Society (EMBC)},
  year={2020}
}

---

Authors

This work is conducted by Hong Kong University of Science and Technology (HKUST), Southern University of Science and Technology (SUSTech), WeBank Co. Ltd. and Nanyang Technology University (NTU).

The authors are