Table 4.

Imaging site prediction results with FD simulated data.

k	n	External test set	Leaked internal test set	Not leaked internal test set
3	25	0.358 (0.078)	0.288 (0.048)*	0.369 (0.046)
3	50	0.324 (0.060)	0.270 (0.045)*	0.333 (0.045)
3	100	0.352 (0.047)	0.296 (0.029)*	0.347 (0.027)
3	250	0.328 (0.021)	0.292 (0.019)*	0.325 (0.017)
10	25	0.084 (0.026)	0.049 (0.014)*	0.089 (0.017)
10	50	0.104 (0.017)	0.068 (0.011)*	0.105 (0.013)
10	100	0.098 (0.013)	0.072 (0.008)*	0.097 (0.009)
10	250	0.097 (0.008)	0.084 (0.005)*	0.100 (0.005)
36	25	0.024 (0.007)	0.010 (0.002)*	0.026 (0.004)
36	50	0.028 (0.005)	0.013 (0.002)*	0.026 (0.002)^
36	100	0.028 (0.003)	0.017 (0.002)*	0.028 (0.002)
36	250	0.028 (0.003)	0.021 (0.001)*	0.028 (0.001)

Average balanced accuracy (standard deviation) obtained in the external and internal test sets.

k is the number of single-center datasets, each one containing n participants. FD: fractal dimension.

*One-tailed paired t-test Bonferroni adjusted p-value < 10⁻¹⁰ for the comparison with the external test set average balanced accuracy.

^One-tailed t-test Bonferroni adjusted p-value = 0.003 for the comparison with the external test set average balanced accuracy.