[Retracted] Automatic Detection of High-Frequency Oscillations Based on an End-to-End Bi-Branch Neural Network and Clinical Cross-Validation

<div>Visual marking example. The upper row is the raw data, filtered signal (80–500 Hz) lies in the middle row, and the time-frequency diagram is shown at the bottom row. (a) Ripple; (b) fast ripple; (c) artifact 1, high frequency transients caused by a band-pass filter; (d) artifact 2, the amplitude is larger than the global background but not significantly different from the local activity; (e) artifact 3, harmonics of low frequency nonsinusoidal signals.</div>

Computational Intelligence and Neuroscience

fig3

Figure 3

Figure 3: [Retracted] Automatic Detection of High-Frequency Oscillations Based on an End-to-End Bi-Branch Neural Network and Clinical Cross-Validation 

Figure 3 | [Retracted] Automatic Detection of High-Frequency Oscillations Based on an End-to-End Bi-Branch Neural Network and Clinical Cross-Validation 