TechRxiv
Ahmadi_2020_techRxiv_BAKS_LSTM.pdf (5.28 MB)

Improved Spike-based Brain-Machine Interface Using Bayesian Adaptive Kernel Smoother and Deep Learning

Download (5.28 MB)
preprint
posted on 29.05.2020, 22:04 by Nur Ahmadi, Timothy Constandinou, Christos-Savvas Bouganis
Multiunit activity (MUA) has been proposed to mitigate the robustness issue faced by single-unit activity (SUA)-based brain-machine interfaces (BMIs). Most MUA-based BMIs still employ a binning method for extracting firing rates and linear decoder for decoding behavioural parameters. The limitations of binning and linear decoder lead to suboptimal performance of MUA-based BMIs. To address this issue, we propose Bayesian adaptive kernel smoother (BAKS) as the feature extraction method and long short-term memory (LSTM)-based deep learning as the decoding algorithm. We evaluated the proposed methods for reconstructing (offline) hand kinematics from intracortical neural data chronically recorded from the motor cortex of a monkey. Experimental results showed that BAKS coupled with LSTM outperformed other combinations of feature extraction method (binning or fixed kernel smoother) and decoding algorithm (Kalman filter or Wiener filter). Overall results demonstrate the effectiveness of BAKS and LSTM for improving the decoding performance of MUA-based BMIs.

Funding

Engineering and Physical Sciences Research Council (EPSRC) Award (No. EP/M020975/1)

Indonesia Endowment Fund for Education (LPDP) Award (No. PRJ-123/LPDP/2016)

History

Email Address of Submitting Author

n.ahmadi16@imperial.ac.uk

ORCID of Submitting Author

0000-0002-9902-9051

Submitting Author's Institution

Imperial College London

Submitting Author's Country

United Kingdom

Licence

Exports

Categories

Licence

Exports