Application of Deep Learning on the Prognosis of Cutaneous Melanoma Based on Full Scan Pathology Images

Figure 1

(a) Receiver operating characteristic (ROC) curves and area under the curve (AUC) of the nomogram for the accuracy of deep learning model diagnosis. (b, c) Plots for lasso regression coefficients over different values of the deep feature from pathology scan images. (d) Receiver operating characteristic (ROC) curves and area under the curve (AUC) of the nomogram for the accuracy of machine learning model diagnosis in the testing set.

(a)

(b)

(c)

(d)