. 2022 Nov 23;12:20151. doi: 10.1038/s41598-022-21085-8

Table 2.

Binary logistic regression model to predict ASD with optimal outcome in 4-years follow-up with 27 main fiber bundles.

DTI metrics	Variables	Estimate	Standard error	Wald	p value	Exp (B)	95% CI Exp (B)
DTI metrics	Variables	Estimate	Standard error	Wald	p value	Exp (B)	Lower	Upper
FA total model variance explained, r² = 0.494	str_l	1.83	0.47	15.32	< 0.001***	6.25	2.50	15.63
FA total model variance explained, r² = 0.494	Constant	0.26	0.34	0.61	0.436	1.30	–	–
MD total model variance explained, r² = 0.796	fma	3.35	0.96	12.14	< 0.001***	28.46	4.33	187.10
	str_l	− 2.30	0.71	10.62	0.001**	0.10	0.03	0.40
	Constant	0.50	0.55	0.82	0.364	1.64	–	–
AD total model variance explained, r² = 0.643	ar_l	1.86	0.82	5.17	0.023*	6.40	1.29	31.67
	atr_l	3.60	1.11	10.59	0.001**	36.43	4.18	317.61
	str_l	− 4.13	1.17	12.41	< 0.001***	0.02	0.002	0.16
	Constant	− 0.14	0.43	0.10	0.752	0.87	–	–
RD total model variance explained, r² = 0.74	str_l	− 1.75	0.49	12.86	< 0.001***	0.17	0.07	0.45
RD total model variance explained, r² = 0.74	Constant	− 0.18	0.32	0.33	0.564	0.83	–	–

*p < 0.05; **p < 0.01; ***p < 0.001. FA fractional anisotropy, MD mean diffusivity, AD axial diffusivity, RD radial diffusivity, str_l the left superior longitudinal fasciculus, fma forceps major, ar_l the left acoustic radiation, atr_l the left anterior thalamic radiation.