Fig. 4: Development of a clinical signature (CS)-based platform for predicting cancer patient survival using Cox regression analyses. | npj Precision Oncology

Fig. 4: Development of a clinical signature (CS)-based platform for predicting cancer patient survival using Cox regression analyses.

From: Tumor cell- and infiltrating immune cell-based supervised learning artificial intelligence multimodal platform for tumor prognosis

Fig. 4

Univariate (a) and multivariate (b) Cox regression analyses were performed on clinicopathological parameters to evaluate overall survival in patients with oral squamous cell carcinoma (OSCC). Kaplan–Meier (KM) analyses demonstrated moderate predictive accuracy for the CS in the training cohort (c) and validation cohort (d). Based on the median of the expectation of overall survival predicted from the constructed CS platform, OSCC patients were stratified into high-risk (overall survival predicted below the median) and low-risk groups (overall survival predicted above the median). F female, M male, FOM floor of mouth, T stage Tumor stage, N stage Lymph Node stage, Path grade pathological grade, WBC white blood cell, RBC red blood cell, NL neutrophil/lymphocyte ratio, AG albumin/globulin ratio, PL platelet/lymphocyte ratio, HR hazard ratio, CI confidence interval, c-index concordance index.

Back to article page