Decoding methods for neural prostheses: where have we reached?
1State Key Laboratory of Cognitive Neuroscience and Learning and IDG/McGovern Institute for Brain Research, Beijing Normal University Beijing, China ; Center for Collaboration and Innovation in Brain and Learning Sciences, Beijing Normal University Beijing, China.
Frontiers in Systems Neuroscience
|August 1, 2014
Summary
This review covers brain-machine interface (BMI) decoding methods, focusing on practical challenges for prosthetic deployment. It addresses key questions for improving control accuracy and adaptability in neural prosthetics.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Rehabilitation Technology
Background:
- Brain-machine interfaces (BMIs) offer potential for restoring function through neural prosthetics.
- Clinical deployment of BMIs necessitates practical and robust decoding methods.
Purpose of the Study:
- To review recent advances in decoding methods for brain-machine interfaces.
- To identify and discuss open questions critical for the clinical application of prosthetic devices.
Main Methods:
- Review of current literature on BMI decoding techniques.
- Organization of findings around key challenges in decoder design and implementation.
Main Results:
- Advances in decoding methods are reviewed, focusing on practical deployment considerations.
- Key challenges include variable decoding, neural tuning models, neuron selection, action modeling, parameter learning, and signal adaptation.
Conclusions:
- Decoders must be designed and tested within the context of their intended use.
- Determining adequate control accuracy for prosthetic function remains a critical open question.
Keywords:
brain computer interfacebrain-machine interfacedecodingmultichannel recordingsneural engineeringneural prostheticsignal processing
