Impact of a machine learning–based prediction model on annual surveillance endoscopy costs for detecting gastric cancer

Background and Aims

In this study, we assessed our machine learning (ML)-based model's impact on reducing annual surveillance endoscopy costs for detecting gastric cancer (GC).

Methods

We analyzed 1099 patients with chronic gastritis undergoing annual EGD and randomly divided them into training and test sets (4:1). Using gradient-boosting decision trees and incorporating patient characteristics, we developed the ML model. In the test sets, we compared the EGD number needed to screen (NNS) for 1 GC, cost, and GC detection rate across different risk stratification strategies.

Results

The ML-selected high-risk cohort demonstrated low NNS values, low total cost, low cost per 1 GC, and high GC detection rates compared with alternative risk stratification approaches, including operative link for gastric atrophy assessment and operative link for gastric intestinal metaplasia assessment.

Conclusions

Our ML model holds promise in reducing endoscopy surveillance costs while maintaining a robust GC detection rate.