Name
Abstract | ||
|---|---|---|
Brain-machine interfaces (BMIs) use neural activity related to motion parameters to enable brain directly control external devices. Some linear and nonlinear decoding techniques have been used successfully to infer arm trajectory from neural data. Unfortunately, these One stage decoding techniques can hardly get high accuracy and low computational demands at the same time. Here we introduce a Two Stage Model (TSM) which consists of two linear models, on the basis that different motion states have different neural firing patterns when rats were doing the lever pressing task. The accuracies of the neural firing patterns classification were higher than 90% for all the three datasets. The Correlation coefficients (CC) between the trajectory predicted by TSM and the measured one were up to 0.89, 0.85 and 0.95 for the three datasets respectively higher than those of Kalman Filter (KF) and Partial Least Squares Regression (PLSR). The time consumption of TSM was about only 10% of that of Generalized Regression Neural Network (GRNN). These results show that TSM can simultaneously get both high accuracy and low computational cost. |
Year | DOI | Venue |
|---|---|---|
2011 | 10.1109/IEMBS.2011.6090627 | 2011 ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY (EMBC) |
Keywords | Field | DocType |
brain machine interface,neural nets,brain computer interfaces,decoding,regression analysis,kalman filters,kalman filter,partial least square regression,classification algorithms,linear model,accuracy | Regression analysis,Computer science,Partial least squares regression,Artificial intelligence,Artificial neural network,Trajectory,Computer vision,Pattern recognition,Linear model,Kalman filter,Decoding methods,Statistical classification,Machine learning | Conference |
Volume | Issue | ISSN |
2011 | null | 1557-170X |
Citations | PageRank | References |
0 | 0.34 | 1 |
Authors | ||
6 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Bo Jiang | 1 | 0 | 0.34 |
| Rui Wang | 2 | 489 | 33.21 |
| Qiaosheng Zhang | 3 | 24 | 7.54 |
| Jicai Zhang | 4 | 0 | 0.34 |
| Xiaoxiang Zheng | 5 | 125 | 27.26 |
| Ting Zhao | 6 | 5 | 5.22 |