Cancer Discrimination Using Fourier Transform Near-Infrared Spectroscopy with Chemometric Models

Table 1

Summarization of model performance of different classification models.

Preprocessing method	Training			Test			Parameter		ROC curve
	MCR	SENS	SPEC	MCR	SENS	SPEC	Gam	Sig2	Area	Std
Original	0.16	0.90	0.71	0.18	0.81	0.81	0.95	165	0.94	0.02
SNV	0	1.00	1.00	0.08	0.94	0.85	34.18	105	1	0
Derivative	0	1.00	1.00	0.32	1.00	0.03	0.69	13.3	1	0
SNV-PC	0.28	0.98	0.21	0.29	0.96	0.18	0.94	1.60	0.75	0.06
Derivative-PC	0.04	0.96	0.96	0.05	0.94	0.96	7.30	4.66	0.99	0.004