Table 1 Age modeling from structural imaging variables.

From: The pediatric template of brain perfusion

 

Estimate

Std. Error

t value

Pr(>|t|)

(Intercept)

12.3390

4.6227

2.67

0.0091

BStem

0.0002

0.0001

3.01

0.0034

CBF

−0.0319

0.0118

−2.71

0.0081

FA

74.6610

10.9782

6.80

0.0000

Thickness

−6.0487

1.0827

−5.59

0.0000

DGM

−0.0001

0.0000

−3.62

0.0005

  1. Cross-validation shows that this simple general linear model can predict age with accuracy of 1.61+/−0.15 years when training on half of the data and testing on the other half, as assessed by 100 data resamplings. Using multiple modalities to predict age shows that each modality contributes useful information even in the presence of other imaging measurements.
Back to article page