Copyright: ©Author(s) 2026.
World J Gastroenterol. Sep 14, 2026; 32(34): 118337
Published online Sep 14, 2026. doi: 10.3748/wjg.118337
Published online Sep 14, 2026. doi: 10.3748/wjg.118337
Table 1 Clinical applicability comparison (machine learning models vs traditional markers)
| Model category | Internal AUC | External AUC | Overall performance | Clinical applicability |
| CatBoost (best ML model) | 0.818 | 0.815 | Excellent | Highly recommended |
| Other ML models (average) | 0.792 | 0.79 | Good | Recommended |
| D-dimer (traditional marker) | 0.618 | 0.51 | Poor | Not recommended |
- Citation: Su Y, Wang DX, Zhao YQ, Xing X. Bid farewell to single indicators: Machine learning models integrating multidimensional data lead thrombosis risk prediction into a new stage. World J Gastroenterol 2026; 32(34): 118337
- URL: https://www.wjgnet.com/1007-9327/full/v32/i34/118337.htm
- DOI: https://dx.doi.org/10.3748/wjg.118337