Table 9.

Performance metrics for the best ICU vs FU CNN model found by the optimiser for different sizes of the STFT.

FE domain	Model name	Accuracy	Macro precision	Macro recall	Macro F1	Weighted F1
STFT	CNN–1	0.760	0.776	0.744	0.744	0.751
	CNN–1/2	0.791	0.798	0.783	0.785	0.789
	CNN–1/4	0.827	0.833	0.823	0.824	0.826
	CNN–1/8	0.892	0.891	0.891	0.891	0.892

Size is represented by the number by which the STFT window length is multiplied (e.g. 1/4 represents a quarter of the normal 10-wavelength-long window).