Multi-clinical index classifier combined with AI algorithm model to predict the prognosis of gallbladder cancer

• DOI: 10.3389/fonc.2023.1171837
• Publication Date: 2023-05-10T18:45:50Z
• Authors: Yun Zhou; Siyu Chen; Yuchen Wu; Lanqing Li; Qinqin Lou; Yongyi Chen; Songxiao Xu
• Keywords: Gallbladder Cancer

Objectives It is significant to develop effective prognostic strategies and techniques for improving the survival rate of gallbladder carcinoma (GBC). We aim prediction model from multi-clinical indicators combined artificial intelligence (AI) algorithm prognosis GBC. Methods A total 122 patients with GBC January 2015 December 2019 were collected in this study. Based on analysis correlation, relative risk, receiver operator characteristic curve, importance by AI between clinical factors recurrence survival, two multi-index classifiers (MIC1 MIC2) obtained. The eight algorithms survival. models highest area under curve (AUC) selected test performance testing dataset. Results MIC1 has ten indicators, MIC2 nine indicators. combination classifier “avNNet” can predict an AUC 0.944. “glmet” 0.882. Kaplan-Meier shows that effectively median DFS OS, there no statistically difference results (MIC1: χ 2 = 6.849, P 0.653; MIC2: 9.14, 0.519). Conclusions avNNet mda have high sensitivity specificity predicting