Decoding movements from cortical ensemble activity using a long short-term memory recurrent network
Tseng P. -., Urpi N. A., Lebedev M. A., Nicolelis M.
Neural Computation
Vol.31, Issue6, P. 1085-1113
Опубликовано: 2019
Тип ресурса: Статья
Аннотация:
Although many real-time neural decoding algorithms have been proposed for brain-machine interface (BMI) applications over the years, an optimal, consensual approach remains elusive. Recent advances in deep learning algorithms provide new opportunities for improving the design of BMI decoders, including the use of recurrent artificial neural networks to decode neuronal ensemble activity in real time. Here, we developed a long-short term memory (LSTM) decoder for extracting movement kinematics from the activity of large (N = 134–402) populations of neurons, sampled simultaneously from multiple cortical areas, in rhesus monkeys performing motor tasks. Recorded regions included primary motor, dorsal premotor, supplementary motor, and primary somatosensory cortical areas. The LSTM’s capacity to retain information for extended periods of time enabled accurate decoding for tasks that required both movements and periods of immobility. Our LSTM algorithm significantly outperformed the state-of-
Ключевые слова:
Brain; Brain computer interface; Decoding; Deep learning; Neurons; Brain machine interface; Directional tuning; Movement kinematics; Neuronal activities; Physiological features; Recurrent artificial neural networks; Recurrent networks; Unscented Kalman Filter; Long short-term memory; animal; brain computer interface; motor cortex; movement (physiology); physiology; rhesus monkey; short term memory; somatosensory cortex; Animals; Brain-Computer Interfaces; Macaca mulatta; Memory, Short-Term; Motor Cortex; Movement; Neural Networks, Computer; Somatosensory Cortex
Язык текста: Английский
ISSN: 1530-888X
Tseng P. -. P.-H.
Urpi N. A.
Lebedev M. A. Mikhail Aleksandrovich 1954-
Nicolelis M.
Ценг П. -. П.-Х.
Урпи Н. А.
Лебедев М. А. Михаил Александрович 1954-
Ниcолелис М.
Decoding movements from cortical ensemble activity using a long short-term memory recurrent network
Текст визуальный непосредственный
Neural Computation
MIT Press
Vol.31, Issue6 P. 1085-1113
2019
Статья
Brain Brain computer interface Decoding Deep learning Neurons Brain machine interface Directional tuning Movement kinematics Neuronal activities Physiological features Recurrent artificial neural networks Recurrent networks Unscented Kalman Filter Long short-term memory animal brain computer interface motor cortex movement (physiology) physiology rhesus monkey short term memory somatosensory cortex Animals Brain-Computer Interfaces Macaca mulatta Memory, Short-Term Motor Cortex Movement Neural Networks, Computer Somatosensory Cortex
Although many real-time neural decoding algorithms have been proposed for brain-machine interface (BMI) applications over the years, an optimal, consensual approach remains elusive. Recent advances in deep learning algorithms provide new opportunities for improving the design of BMI decoders, including the use of recurrent artificial neural networks to decode neuronal ensemble activity in real time. Here, we developed a long-short term memory (LSTM) decoder for extracting movement kinematics from the activity of large (N = 134–402) populations of neurons, sampled simultaneously from multiple cortical areas, in rhesus monkeys performing motor tasks. Recorded regions included primary motor, dorsal premotor, supplementary motor, and primary somatosensory cortical areas. The LSTM’s capacity to retain information for extended periods of time enabled accurate decoding for tasks that required both movements and periods of immobility. Our LSTM algorithm significantly outperformed the state-of-