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

Table 7

The comparison between proposed method and existing method using cross-subject validation.

Methods	SEN (%)	SPE (%)	PRE (%)	ACC (%)	FDR (%)	F1 (%)	SEN_SPE (%)
[28]	77.04 (R)	72.27 (R)	—	—	—	—	74.58 (R)
	83.23 (FR)	79.36 (FR)					81.25 (FR)
[29]	92.4 (R)	—	—	—	9.2 (R)	—	—
	90.3 (FR)				10.7 (FR)
[47]	82.666	63.352	—	—	—	—	71.73
Proposed	92.00	88.26	86.86	89.76	13.14	89.11	89.87