Construction and implications of structural equation modeling network for pediatric cataract: a data mining research of rare diseases

Table 2 Summary of the distribution of the 8 filtered variables included in the final SEM analysis

Two-layer Variables	One-layer Variables	Original Variables	Detailed distribution
Overall index	Concomitant variables	Age at diagnosis	50.99 ± 36.38 months
		Abnormal pregnancy history	23.13% (+)
		Abnormal pregnancy history	76.87% (−)
	Structural indices	AL	21.75 ± 2.06 mm
		Area	55% (Extensive)
		Area	45% (Limited)
		Density	37.5% (Dense)
		Density	63.5% (Non-dense)
		Location	60.63% (Central)
		Location	39.37% (Peripheral)
	Functional indices	IOP	15.12 ± 6.88 mmHg
	Functional indices	UCVA	0.77 ± 0.44 (logMAR)

A total of 160 participants from the CCPMOH database and 8 filtered variables were included in the final SEM analysis. The mean age of the included participants was 50.99 months ± 36.38 months, and 23.13% (n = 37) of our patients had an abnormal pregnancy history. For the STR network, the mean value of AL was 21.75 mm ± 2.06 mm; 55% (n = 88) of our patients had an extensive area; 37.5% (n = 60) of our patients had dense opacity, and 60.63% (n = 97) of our patients had opacity at central location. For the FUN network, the mean IOP value was 15.12 mmHg ±6.88 mmHg, and the mean UCVA value (logMAR) was 0.77 ± 0.44
AL Axial length, IOP Intraocular pressure, UCVA uncorrected visual acuity

ISSN: 1471-2415