Table 5.

Overall accuracies of raw and preprocessed clinical spectra classification by SVM, RF, and LDA.

Datasets	# classes	Applied to raw datasets				Applied to preprocessed datasets
		Best CNNs	SVM	RF	LDA	SVM	RF	LDA
Canine sarcoma	2	0.98 ± 0.00a	0.77 ± 0.02	0.96 ± 0.01	0.71 ± 0.02	0.76 ± 0.16	0.96 ± 0.01	0.93 ± 0.02
	12	0.99 ± 0.00b	0.61 ± 0.00	0.65 ± 0.04	0.41 ± 0.01	0.52 ± 0.19	0.65 ± 0.01	0.60 ± 0.04
Microorganisms	3	0.99 ± 0.00c	0.45 ± 0.03	0.77 ± 0.03	0.90 ± 0.01	0.87 ± 0.02	0.95 ± 0.02	0.88 ± 0.02
	5	0.99 ± 0.00c	0.54 ± 0.35	0.86 ± 0.01	0.67 ± 0.13	0.19 ± 0.09	0.87 ± 0.02	0.85 ± 0.03
Human ovary 1	2	0.98 ± 0.00c	0.53 ± 0.04	0.84 ± 0.05	0.65 ± 0.04	0.66 ± 0.24	0.91 ± 0.02	0.93 ± 0.02
Human ovary 2	2	0.99  ± 0.00b	0.60 ± 0.06	0.81 ± 0.01	0.71 ± 0.03	0.60 ± 0.05	0.88 ± 0.03	0.96 ± 0.00

The best result for each task (accuracy ± standard variation over 10 independent iterations) is indicated in boldface.

^aThe best CNNs from scratch.

^bThe best CNNs after cumulative learning.

^cThe best CNNs after transfer learning.