TABLE 4.
Entorhinal cortex texture and volume in classifying NC vs. AD.
NC vs. AD | ROC analysis AUC | 95% CI | P-value |
Entorhinal cortex | |||
Texture features | |||
ASM | 0.592 | 0.529–0.656 | 0.005 |
Contrast | 0.794 | 0.743–0.845 | <0.001 |
Correlation | 0.824 | 0.776–0.872 | <0.001 |
Variance | 0.524 | 0.458–0.590 | 0.475 |
Sum average | 0.620 | 0.555–0.681 | <0.001 |
Sum variance | 0.713 | 0.653–0.770 | <0.001 |
Entropy | 0.685 | 0.625–0.744 | <0.001 |
Cluster shade | 0.540 | 0.475–0.604 | 0.238 |
Volume and thickness | |||
Erc. volume | 0.888 | 0.847–0.925 | <0.001 |
Erc. thickness | 0.809 | 0.755–0.863 | <0.001 |
Features combination | |||
Texture (ASM, correlation, variance, sum average, and cluster shade) | 0.872 | 0.828–0.916 | <0.001 |
Texture + Erc. volume | 0.914 | 0.879–0.950 | <0.001 |
Hippocampus | |||
Hippocampal volume | 0.869 | 0.827–0.912 | <0.001 |
NC, Normal Controls; AD, Alzheimer’s disease; ROC, receiver operating characteristic; AUC, area under curve, CI, confidence interval; ASM, angular second moment; Erc, entorhinal cortex.