Skip to main content
Fig. 7 | Journal of Biomedical Science

Fig. 7

From: Diagnosis of transition zone prostate cancer by multiparametric MRI: added value of MR spectroscopic imaging with sLASER volume selection

Fig. 7

Results of support vector machine analysis to separate tumor aggressiveness classes. Left side: ROC curves of the six RBF-SVM models for low-risk vs high-risk cancer, low-risk vs intermediate-risk cancer and intermediate-risk vs high-risk cancer. A leave-one-out cross-validation technique was used for the combined ADC and Ktrans (dashed line) and all the combined ADC, Ktrans and metabolite ratios (solid line) with a significant difference between the two groups (p < 0.05). Right side: bar charts of AUC values, sensitivity and specificity of the corresponding RBF-SVM models. McNemar test was used for pairwise comparison of sensitivity and specificities of models and Delong test was calculated for pairwise comparison of AUC of models. **p < 0.01 and * p < 0.05

Back to article page