Research Article
Spike Sorting by Joint Probabilistic Modeling of Neural Spike Trains and Waveforms
Table 3
Classification error rates for the WaveClus semiartificial dataset.
| Dataset | SNR | GMM | WaveClus method | Proposed |
| Easy1 | Clean | 0.93% | 0.00% | 0.97% | 10 dB | 0.72% | 0.00% | 0.77% | 5 dB | 0.89% | 0.00% | 0.85% | 0 dB | 0.81% | 0.21% | 0.85% | −5 dB | 1.25% | 0.52% | 0.97% | −10 dB | 8.18% | 3.47% | 6.20% |
| Difficult1 | Clean | 3.18% | 0.45% | 2.58% | 10 dB | 4.83% | 0.94% | 3.46% | 5 dB | 6.89% | 1.36% | 5.32% | 0 dB | 19.77% | 20.0% | 37.21% |
|
|