. Author manuscript; available in PMC: 2014 Dec 1.

Published in final edited form as: Neuroimage. 2013 Jun 21;83:10.1016/j.neuroimage.2013.06.033. doi: 10.1016/j.neuroimage.2013.06.033

Table 4.

Accuracy of classifications based on linear discriminant analysis with and without LLE feature embeddings

		CN vs. s-MCI			CN vs. c-MCI			CN vs. AD			s-MCI vs. c-MCI			s-MCI vs. AD			c-MCI vs. AD
		ACR	SEN	SPE	ACR	SEN	SPE	ACR	SEN	SPE	ACR	SEN	SPE	ACR	SEN	SPE	ACR	SEN	SPE
Volumes	LDA	0.56^a	0.60	0.54	0.77	0.75	0.78	0.83	0.85	0.82	0.46^a	0.48	0.44	0.58^a	0.66	0.51	0.49^a	0.45	0.53
	LDA+LLE	0.59	0.56	0.61	0.81	0.79	0.83	0.84	0.77	0.89	0.61	0.70	0.62	0.67	0.72	0.62	0.45^a	0.47	0.43
	p-value^b	0.59			0.01			0.14			0.08			0.05			0.88

Cortical thickness	LDA	0.55^a	0.54	0.57	0.73	0.66	0.78	0.84	0.81	0.86	0.57^a	0.54	0.60	0.66	0.80	0.53	0.56^a	0.58	0.55
	LDA+LLE	0.69	0.67	0.71	0.78	0.79	0.77	0.85	0.79	0.77	0.65	0.52	0.56	0.66	0.71	0.61	0.58^a	0.70	0.48
	p-value^b	0.002			0.07			0.61			0.04			0.79			0.66

Thickness +volume	LDA	0.55^a	0.56	0.54	0.67	0.69	0.65	0.69	0.62	0.73	0.51^a	0.57	0.45	0.58^a	0.63	0.54	0.45^a	0.43	0.46
	LDA+LLE	0.51^a	0.48	0.49	0.75	0.69	0.80	0.89	0.86	0.91	0.68	0.82	0.53	0.65	0.70	0.60	0.57^a	0.51	0.63
	p-value^b	0.89			<0.001			<0.001			0.01			0.03			0.15

indicates classification is not better than by chance (p>0.05), whereas no superscripts indicate a significant classification;

p-value of improvement with LLE compared to non-LLE.