Research Article

Constructing the Logical Regression Model to Predict the Target of Jianpi Jiedu Decoction in the Treatment of Hepatocellular Carcinoma

Figure 4

Assessment of the 12-gene prognostic signature and a compound nomogram establishment. (a) Univariate independent prognostic analysis chart, we analyzed age, gender, grade, stages T, M, and N, and risk score. And stage T was the high risk independent prognostic factor (). (b) Multivariate independent prognostic analysis showed that M and risk score were high risk factors (). (c) ROC curve showed that the risk score (AUC = 0.776) had better prediction performance than stage (AUC = 0.665), T (AUC = 0.670), and other factors. (d) Nomogram. According to AJCC stage, T, M, and N and risk factors, the model was constructed by Cox analysis, and the 1-year, 3-year, and 5-year survival rate of patients with HCC was evaluated according to the total score of clinical characteristics.
(a)
(b)
(c)
(d)