An evidential reasoning based model for diagnosis of lymph node metastasis in gastric cancer

Table 4 AUC values of nine methods on testing data

Model	Sensitivity	Specificity	AUC¹	P value (AUC compared with Radiologist)
ER based model	0.8951	0.8	0.829 ± 0.037	P < 0.001
ANN	0.8453	0.6429	0.791 ± 0.041	P < 0.001
SVM Linear	0.8723	0.6765	0.78 ± 0.046	P < 0.001
SVM Gaussion	0.7657	0.6036	0.688 ± 0.059	P < 0.001
SVM Ploynomial2	0.8611	0.6774	0.781 ± 0.048	P < 0.001
SVM Ploynomial3	0.8662	0.6667	0.782 ± 0.048	P < 0.001
SVM Ploynomial4	0.8633	0.6111	0.769 ± 0.049	P < 0.001
Logistic Regression	0.8552	0.6667	0.793 ± 0.039	P < 0.001
Radiologist²	0.634	0.765	0.757 ± 0.042

¹The value of data was AUC + standard deviation.
²Taking 7.7 mm as the best cut-off point of maximum lymph node size, and if lymph node size larger than 7.7 mm, it is considered that LNM occurs; otherwise, it does not occur.

ISSN: 1472-6947