A stack LSTM structure for decoding continuous force from local field potential signal of primary motor cortex (M1)

Background Brain Computer Interfaces (BCIs) translate the activity of the nervous system to a control signal which is interpretable for an external device. Using continuous motor BCIs, the user will be able to control a robotic arm or a disabled limb continuously. In addition to decoding the target position, accurate decoding of force amplitude is essential for designing BCI systems capable of performing fine movements like grasping. In this study, we proposed a stack Long Short-Term Memory (LSTM) neural network which was able to accurately predict the force amplitude applied by three freely moving rats using their Local Field Potential (LFP) signal. Results The performance of the network was compared with the Partial Least Square (PLS) method. The average coefficient of correlation (r) for three rats were 0.67 in PLS and 0.73 in LSTM based network and the coefficient of determination ( $$R^{2}$$ R 2 ) were 0.45 and 0.54 for PLS and LSTM based network, respectively. The network was able to accurately decode the force values without explicitly using time lags in the input features. Additionally, the proposed method was able to predict zero-force values very accurately due to benefiting from an output nonlinearity. Conclusion The proposed stack LSTM structure was able to predict applied force from the LFP signal accurately. In addition to higher accuracy, these results were achieved without explicitly using time lags in input features which can lead to more accurate and faster BCI systems.

[1]  S. Wold.  Cross-Validatory Estimation of the Number of Components in Factor and Principal Components Models , 1978 .

[2]  Arjun K. Bansal,et al.  Relationships among low-frequency local field potentials, spiking activity, and three-dimensional reach and grasp kinematics in primary motor and ventral premotor cortices. , 2011, Journal of neurophysiology.

[3]  Sonja Grün,et al.  Local field potentials in primate motor cortex encode grasp kinetic parameters , 2015, NeuroImage.

[4]  J. Wolpaw,et al.  Brain-Computer Interfaces: Principles and Practice , 2012 .

[5]  Naotaka Fujii,et al.  Decoding continuous three-dimensional hand trajectories from epidural electrocorticographic signals in Japanese macaques , 2012, Journal of neural engineering.

[6]  Yoshua Bengio,et al.  Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.

[7]  John P. Donoghue,et al.  Decoding 3-D Reach and Grasp Kinematics From High-Frequency Local Field Potentials in Primate Primary Motor Cortex , 2010, IEEE Transactions on Biomedical Engineering.

[8]  Bijan Pesaran,et al.  Optimizing the Decoding of Movement Goals from Local Field Potentials in Macaque Cortex , 2011, The Journal of Neuroscience.

[9]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[10]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[11]  Gilles Louppe,et al.  Independent consultant , 2013 .

[12]  Tom Heskes,et al.  On the decoding of intracranial data using sparse orthonormalized partial least squares. , 2012, Journal of neural engineering.

[13]  E. Keefer,et al.  Human motor decoding from neural signals: a review , 2019, BMC biomedical engineering.

[14]  Yasuharu Koike,et al.  Decoding grasp force profile from electrocorticography signals in non-human primate sensorimotor cortex , 2014, Neuroscience Research.

[15]  J. Donoghue,et al.  Primary Motor Cortex Tuning to Intended Movement Kinematics in Humans with Tetraplegia , 2008, The Journal of Neuroscience.

[16]  Mohammad Reza Daliri,et al.  Adaptive Artifact Removal From Intracortical Channels for Accurate Decoding of a Force Signal in Freely Moving Rats , 2019, Front. Neurosci..

[17]  M. Daliri,et al.  Continuous Force Decoding from Local Field Potentials of the Primary Motor Cortex in Freely Moving Rats , 2016, Scientific Reports.

[18]  Simin Li,et al.  An Improved Unscented Kalman Filter Based Decoder for Cortical Brain-Machine Interfaces , 2016, Frontiers in neuroscience.

[19]  Mohammad Reza Daliri,et al.  Minimum Noise Estimate Filter: A Novel Automated Artifacts Removal Method for Field Potentials , 2017, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[20]  Elaine B. Martin,et al.  Model selection for partial least squares regression , 2002 .

[21]  A R Marathe,et al.  Decoding continuous limb movements from high-density epidural electrode arrays using custom spatial filters , 2013, Journal of neural engineering.

[22]  Zoran Nenadic,et al.  Extracting kinetic information from human motor cortical signals , 2014, NeuroImage.

[23]  David D. Cox,et al.  Making a Science of Model Search: Hyperparameter Optimization in Hundreds of Dimensions for Vision Architectures , 2013, ICML.

[24]  Joseph E. O’Doherty,et al.  Unscented Kalman Filter for Brain-Machine Interfaces , 2009, PloS one.

[25]  Jürgen Schmidhuber,et al.  LSTM: A Search Space Odyssey , 2015, IEEE Transactions on Neural Networks and Learning Systems.

[26]  S. Acharya,et al.  Connectivity Analysis as a Novel Approach to Motor Decoding for Prosthesis Control , 2012, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[27]  Hugo Gamboa,et al.  Biosignals learning and synthesis using deep neural networks , 2017, BioMedical Engineering OnLine.

[28]  Kip A Ludwig,et al.  Using a common average reference to improve cortical neuron recordings from microelectrode arrays. , 2009, Journal of neurophysiology.

[29]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[30]  Timothy G. Constandinou,et al.  Decoding Hand Kinematics from Local Field Potentials Using Long Short-Term Memory (LSTM) Network , 2019, 2019 9th International IEEE/EMBS Conference on Neural Engineering (NER).

[31]  Jon A. Mukand,et al.  Neuronal ensemble control of prosthetic devices by a human with tetraplegia , 2006, Nature.

[32]  R. Andersen,et al.  Selecting the signals for a brain–machine interface , 2004, Current Opinion in Neurobiology.