Development and validation of an interpretable machine learning model for predicting chronic atrophic gastritis in elderly patients

1

Chronic atrophic gastritis (CAG) is a significant precancerous condition, particularly prevalent in the elderly population.
2

The study developed and validated interpretable machine learning models to predict CAG in elderly patients using two independent cohorts.
3

Eight key predictive variables were identified through multi-dimensional feature selection methods for the machine learning models.
4

The Multilayer Perceptron model showed optimal performance with an AUC of 0.826 in internal validation and 0.780 in external validation.
5

SHAP analysis indicated that Helicobacter pylori infection, age, smoking, and high-salt food intake are major risk factors for CAG.

Frontiers In Medicine

by Wenjing Fan, Jinyu Wang, Lu Li, Chao Tian, Zhiwei Yang, Guangchao Zhang, Deyu Xu, Xingtang Yang
July 2, 2026

1