Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Superiority of nonlinear mapping in decoding multiple single-unit neuronal spike trains: a simulation study.

Kyung Hwan Kim¹, Sung Shin Kim, Sung June Kim

¹Department of Biomedical Engineering, College of Health Science, Yonsei University, Wonju, Kangwon-do, South Korea. khkim@dragon.yonsei.ac.kr

Journal of Neuroscience Methods

|August 16, 2005

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Endogenous Thrombin Potential Level Helps Predict High Blood Loss in Patients Undergoing Cardiac Surgery.

Annals of laboratory medicine·2024

Same author

Unraveling the Impact of Sarcopenia-Induced Lymphopenia on Treatment Response and Prognosis in Patients with Stage III Non-Small Cell Lung Cancer: Insights for Optimizing Chemoradiation and Immune Checkpoint Inhibitor.

Cancer research and treatment·2024

Same author

Costs Associated with Transcatheter Aortic Valve Implantation and Surgical Aortic Valve Replacement in Korea.

Journal of chest surgery·2024

Same author

Impact of diabetes mellitus and hypertension on renal function during first-line targeted therapy for metastatic renal cell carcinoma: a retrospective multicenter study.

Translational andrology and urology·2024

Same author

Characteristics of secondary aerosol formation during shortened multiday reaction experiments in a smog chamber: Effects of relative humidity and ammonia.

The Science of the total environment·2024

Same author

Development of learning-based predictive models for radiation-induced atrial fibrillation in non-small cell lung cancer patients by integrating patient-specific clinical, dosimetry, and diagnostic information.

Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology·2024

Nonlinear decoding algorithms, particularly Support Vector Machine Regression (SVR), outperform linear filters for brain-machine interfaces (BMI). SVR demonstrates superior performance in reconstructing neural information from spike trains, especially when dealing with errors.

Area of Science:

Neuroscience
Computational Neuroscience
Machine Learning

Background:

Decoding algorithms are crucial for brain-machine interfaces (BMI) that use neuronal spike trains.
Previous research suggested linear filters are sufficient, with minimal gains from nonlinear methods.

Purpose of the Study:

To investigate the efficacy of nonlinear decoding algorithms compared to linear filters for BMI.
To evaluate multilayer perceptron (MLP) and support vector machine regression (SVR) for spike train decoding.

Main Methods:

Developed linear filter-based and nonlinear (MLP, SVR) decoding algorithms.
Assessed algorithm performance using spike trains from biophysical neuron and primary motor cortex models.

Main Results:

Related Experiment Videos

Nonlinear algorithms generally showed superior performance over linear filters.
The benefits of nonlinear methods were more pronounced with false-positive/negative errors in spike trains.
Support Vector Machine Regression (SVR) achieved the highest decoding performance.

Conclusions:

Nonlinear decoding algorithms, especially SVR, offer significant advantages for brain-machine interfaces.
SVR's superior training and generalization capabilities contribute to its high performance in neural decoding.