©The Author(s) 2026.
World J Gastrointest Oncol. Feb 15, 2026; 18(2): 113959
Published online Feb 15, 2026. doi: 10.4251/wjgo.v18.i2.113959
Published online Feb 15, 2026. doi: 10.4251/wjgo.v18.i2.113959
Table 3 Model comparison
| Data | Model | AUC (95%CI) | Sensitivity (95%CI) | Specificity (95%CI) |
| Train | Random forest | 0.8542 (0.8307-0.8777) | 0.7349 (0.6918-0.774) | 0.8127 (0.7746-0.8459) |
| Logistic | 0.8208 (0.7940-0.8475) | 0.7284 (0.6889-0.7679) | 0.7802 (0.7425-0.8179) | |
| Test | Random forest | 0.8073 (0.759-0.8557) | 0.7226 (0.644-0.7899) | 0.7853 (0.7127-0.844) |
| Logistic | 0.8064 (0.7587-0.8540) | 0.7178 (0.6487-0.7869) | 0.7806 (0.7155-0.8458) |
- Citation: Cao H, Han JL, Wu H, Si SP, Ding LJ, Ji L, Zhang HZ, Yin J, Zhou ZY, Zhang YN, Lv ZF, Tian WY, Zhan Q, Wang H, An FM. Risk prediction for chronic atrophic gastritis using a random forest model: A multicenter study. World J Gastrointest Oncol 2026; 18(2): 113959
- URL: https://www.wjgnet.com/1948-5204/full/v18/i2/113959.htm
- DOI: https://dx.doi.org/10.4251/wjgo.v18.i2.113959