We present neural decoding results for both single and multi-finger movements depending on neural subsets. Experimentally, data were collected from 115 task-related neurons in M1 as the monkey preferred flexion and extension of each finger and the wrist (12 single and 6 multi-movements). The neural decoding is done by an optimal method which is based on the maximum likelihood (ML) inference. Each neuron's activation is quantified by the change in firing rate in before and after finger movements. The results show that with as few as 20-25 randomly selected neurons, we achieved 99% or higher decoding accuracy for single finger movements. The decoding accuracy was 5-10% lower for two-finger movements, but increased to greater than 95% with 30 or more neurons.
|Original language||English (US)|
|Number of pages||4|
|Journal||World Academy of Science, Engineering and Technology|
|State||Published - Nov 2009|
ASJC Scopus subject areas