TABLE 7.
Entorhinal cortex texture and volume in classifying MCI vs. AD.
MCI vs. AD | ROC analysis AUC | 95% CI | P-value |
Entorhinal cortex | |||
Texture features | |||
ASM | 0.627 | 0.565–0.690 | <0.001 |
Contrast | 0.704 | 0.646–0.763 | <0.001 |
Correlation | 0.725 | 0.668–0.783 | <0.001 |
Variance | 0.624 | 0.560–0.688 | <0.001 |
Sum average | 0.649 | 0.587–0.711 | <0.001 |
Sum variance | 0.658 | 0.596–0.720 | <0.001 |
Entropy | 0.656 | 0.594–0.718 | <0.001 |
Cluster shade | 0.645 | 0.583–0.706 | <0.001 |
Volume and thickness | |||
Erc. volume | 0.726 | 0.670–0.781 | <0.001 |
Erc. thickness | 0.702 | 0.642–0.762 | <0.001 |
Features combination | |||
Texture (ASM, correlation, variance, sum average, and cluster shade) | 0.764 | 0.710–0.818 | <0.001 |
Texture + Erc. volume | 0.780 | 0.728–0.833 | <0.001 |
Hippocampus | |||
Hippocampal volume | 0.711 | 0.652–0.771 | <0.001 |
MCI, mild cognitive impairment; AD, Alzheimer’s disease; ROC, receiver operating characteristic; AUC, area under curve, CI, confidence interval; ASM, angular second moment; Erc, entorhinal cortex.