TY - JOUR
T1 - Deep Pinsker and James-Stein Neural Networks for Decoding Motor Intentions from Limited Data
AU - Angjelichinoski, Marko
AU - Soltani, Mohammadreza
AU - Choi, John
AU - Pesaran, Bijan
AU - Tarokh, Vahid
N1 - Funding Information:
Manuscript received September 24, 2020; revised April 1, 2021; accepted May 16, 2021. Date of publication May 26, 2021; date of current version June 11, 2021. This work was supported by the Army Research Office MURI Contract Number W911NF-16-1-0368 and DURIP Contract W911NF-21-1-0080. (Corresponding author: Marko Angjelichinoski.) Marko Angjelichinoski, Mohammadreza Soltani, and Vahid Tarokh are with the Department of Electrical and Computer Engineering, Duke University, Durham, NC 27708 USA (e-mail: ma393@duke.edu).
Publisher Copyright:
© 2001-2011 IEEE.
PY - 2021
Y1 - 2021
N2 - Non-parametric regression has been shown to be useful in extracting relevant features from Local Field Potential (LFP) signals for decoding motor intentions. Yet, in many instances, brain-computer interfaces (BCIs) rely on simple classification methods, circumventing deep neural networks (DNNs) due to limited training data. This paper leverages the robustness of several important results in non-parametric regression to harness the potentials of deep learning in limited data setups. We consider a solution that combines Pinsker's theorem as well as its adaptively optimal counterpart due to James-Stein for feature extraction from LFPs, followed by a DNN for classifying motor intentions. We apply our approach to the problem of decoding eye movement intentions from LFPs collected in macaque cortex while the animals perform memory-guided visual saccades to one of eight target locations. The results demonstrate that a DNN classifier trained over the Pinsker features outperforms the benchmark method based on linear discriminant analysis (LDA) trained over the same features.
AB - Non-parametric regression has been shown to be useful in extracting relevant features from Local Field Potential (LFP) signals for decoding motor intentions. Yet, in many instances, brain-computer interfaces (BCIs) rely on simple classification methods, circumventing deep neural networks (DNNs) due to limited training data. This paper leverages the robustness of several important results in non-parametric regression to harness the potentials of deep learning in limited data setups. We consider a solution that combines Pinsker's theorem as well as its adaptively optimal counterpart due to James-Stein for feature extraction from LFPs, followed by a DNN for classifying motor intentions. We apply our approach to the problem of decoding eye movement intentions from LFPs collected in macaque cortex while the animals perform memory-guided visual saccades to one of eight target locations. The results demonstrate that a DNN classifier trained over the Pinsker features outperforms the benchmark method based on linear discriminant analysis (LDA) trained over the same features.
KW - Neural decoding
KW - brain-computer interfaces
KW - local field potentials
KW - neural networks
KW - non-parametric regression
UR - http://www.scopus.com/inward/record.url?scp=85107212099&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85107212099&partnerID=8YFLogxK
U2 - 10.1109/TNSRE.2021.3083755
DO - 10.1109/TNSRE.2021.3083755
M3 - Article
C2 - 34038363
AN - SCOPUS:85107212099
VL - 29
SP - 1058
EP - 1067
JO - IEEE Transactions on Neural Systems and Rehabilitation Engineering
JF - IEEE Transactions on Neural Systems and Rehabilitation Engineering
SN - 1534-4320
M1 - 9440925
ER -