Fig. 8From: Deep learning classification of uveal melanoma based on histopathological images and identification of a novel indicator for prognosis of patientsConstruction of nomogram. AÂ Forest plots of univariate Cox regression for histopathologic DL-signature, gene-signature, and clinical variables. BÂ Forest plots of multivariate Cox regression for histopathologic DL-signature, gene-signature, and clinical variables. CÂ Time dependent C-index of DL-signature, gene-signature, and clinical variables (tumor stage, histological type, metastasis status, chromosome 3 status and SCNA_Cluster). DÂ Nomogram included DL-signature, gene-signature, age, tumor stage and metastasis status which predicts the 3-, and 5-year of overall survival time for patients TCGA-UVM cohort. EÂ Calibration curves of the nomogram for the estimation of 3-, and 5-year of overall survival rates. FÂ Time dependent ROC of nomogram. GÂ The decision curves of nomogram, DL-signature, and gene-signature. HÂ RMS curves for DL-signature and nomogram. IÂ RMS curves for gene-signature and nomogram. JÂ RMS curves for DL-signature and gene-signature. Each point represents the RMS time of corresponding DL-signature, gene-signature and nomogram scoresBack to article page