Table 1.

Performance results of all methods for CRC LNM.

	ACC	AUC	Sensitivity	Specificity	PPV	NPV
AB+GLH	0.6369	0.6357	0.6753	0.6431	0.6910	0.5805
AB+GLCM	0.6458	0.6449	0.6431	0.6491	0.6880	0.6018
AB+SIFT	0.6280	0.6270	0.6267	0.6295	0.6706	0.5836
DT+GLH	0.4598	0.4588	0.4728	0.4441	0.5073	0.4103
DT+GLCM	0.4866	0.4859	0.4972	0.4745	0.5190	0.4529
DT+SIFT	0.4955	0.4950	0.5057	0.4844	0.5190	0.4711
KNN+GLH	0.5967	0.5900	0.5650	0.7458	0.7125	0.2675
KNN+GLCM	0.5818	0.5752	0.5562	0.7	0.7051	0.2553
KNN+SIFT	0.6280	0.6222	0.5886	0.7687	0.7009	0.3435
LR+GLH	0.6429	0.6416	0.6359	0.6519	0.7026	0.5805
LR+GLCM	0.6815	0.6802	0.6684	0.6990	0.7464	0.6140
LR+SIFT	0.6250	0.6242	0.6253	0.6246	0.6618	0.5866
MLP+GLH	0.5402	0.5395	0.5472	0.5321	0.5743	0.5046
MLP+GLCM	0.5759	0.5748	0.5780	0.5733	0.6268	0.5228
MLP+SIFT	0.5744	0.5738	0.5803	0.5678	0.6006	0.5471
NB+GLH	0.6354	0.6331	0.6184	0.6628	0.7464	0.5198
NB+GLCM	0.6637	0.6623	0.6527	0.6782	0.7289	0.5957
NB+SIFT	0.6518	0.6500	0.6373	0.6727	0.7376	0.5623
SGD+GLH	0.6101	0.6089	0.6181	0.6128	0.6742	0.5532
SGD+GLCM	0.5372	0.5375	0.5488	0.5262	0.5248	0.5502
SGD+SIFT	0.5833	0.5838	0.5981	0.5698	0.5598	0.6079
SVM+GLH	0.4896	0.4924	0.5000	0.4836	0.3586	0.6261
SVM+GLCM	0.5208	0.5247	0.5500	0.5076	0.3382	0.7112
SVM+SIFT	0.5327	0.5359	0.5617	0.5172	0.3848	0.6869
LeNet	0.6577	0.7305	0.6535	0.6535	0.6535	0.6535
AlexNet	0.6716	0.7696	0.6708	0.6711	0.6714	0.6706
AlexNet pre-trained model	0.7583	0.7941	0.8004	0.7997	0.7992	0.8009