Table 2.
Results of linear mixed model analysis.
| Dependent variable: | ||||||
|---|---|---|---|---|---|---|
| FA | ADC | FD | FA Peritumoral | ADC Peritumoral | FD Peritumoral | |
| Pathologic hemispheres |
−0.042 | 0.0001 | −0.046 | −0.075 | 0.0001 | −0.067 |
| (−0.047, −0.038) | (0.00005, 0.0001) | (−0.052, −0.039) | (−0.082, −0.069) | (0.0001, 0.0001) | (−0.076, −0.057) | |
| p < 2e-16 | p < 2e-16 | p < 2e-16 | p < 2e-16 | p < 2e-16 | p < 2e-16 | |
| Constant | 0.560 | 0.001 | 0.718 | 0.540 | 0.001 | 0.729 |
| (0.552, 0.567) | (0.001, 0.001) | (0.704, 0.732) | (0.515, 0.564) | (0.001, 0.001) | (0.689, 0.769) | |
| p < 2e-16 | p < 2e-16 | p < 2e-16 | p < 2e-16 | p < 2e-16 | p < 2e-16 | |
| Observations | 13,000 | 13,000 | 13,000 | 4,138 | 4,138 | 4,138 |
| Log Likelihood | 7,926.707 | 97,226.680 | 2,486.618 | 3,201.497 | 31,060.070 | 1,495.331 |
| Akaike Inf. Crit. | −15,845.410 | −194,445.400 | −4,965.236 | −6,394.995 | −62,112.150 | −2,982.661 |
| Bayesian Inf. Crit. | −15,815.520 | −194,415.500 | −4,935.345 | −6,369.683 | −62,086.840 | −2,957.349 |
Models 1–3 show results for the entire CST for FA, ADC, and FD, models 4–6 for the peritumoral segments respectively. The table shows regression coefficients for the fixed effect of hemisphere and the intercept with their respective standard error in brackets. Further, number of observations for each model, the log likelihood ratio, Akaike information criterion, and Bayesian information criterion are stated.