Table 1.

Input normalization in the three artificial neural networks

Categories	Data type		Criterion
Age (yr)	Integral		MMN
Treatment stage	Binary	First-time	0
		Refinement	1
Crowding (mm)	Discrete	No	0
		I°	1/3
		II°	2/3
		III°	1
With/without extraction	Binary	Yes	1
		No	0
Number of extractions	Integral		MMN
Wearing aligners and bonding attachments simultaneously or separately	Binary	Yes	1
		No	0
With/without molar distalization	Binary	Yes	1
		No	0
With/without elastics	Binary	Yes	1
		No	0
Number of elastics	Integral		MMN
With/without interproximal reduction	Binary	Yes	1
		No	0
Amount of interproximal reduction (mm)	Continuous		MMN
Number of teeth with attachments	Integral		MMN
Number of teeth with optimized attachments	Integral		MMN
Number of teeth with lingual attachments	Integral		MMN
Number of upper incisors with attachments	Integral		MMN
Number of lingual buttons	Integral		MMN
With/without precision cut	Binary	Yes	1
		No	0

The crowding data classification: I°, 0 mm ≤ crowding < 4 mm; II°, 4 mm ≤ crowding < 8 mm; and III°, crowding ≥ 8 mm.

MMN, maximum minimum normalization.