Table 1 List of discriminative models.

From: Prediction of future gastric cancer risk using a machine learning algorithm and comprehensive medical check-up data: A case-control study

Models

Classifier

Input variables

Model A

XGBoost

H. pyloria serology testing

Model B

XGBoost

H. pyloria serology testing and chronic atrophic gastritis

Model C

XGBoost

Variables in model B, gastric or duodenal ulcers including scars, GERDb or Barrett’s oesophagus and post-gastrectomy

Model D

XGBoost

Variables in model C, sex, age and body mass index

Model E

XGBoost

Variables in model D, white blood cell counts, neutrophil ratio, lymphocyte ratio, eosinophil ratio, monocyte ratio, basophil ratio, platelet count, haemoglobin, mean corpuscular volume and haemoglobin A1c

Model F

LRc

The same variables as model A

Model G

LR

The same variables as model B

Model H

LR

The same variables as model C

Model I

LR

The same variables as model D

Model J

LR

The same variables as model E

  1. aHelicobacter pylori, H. pylori.
  2. bGastroesophageal reflux disease, GERD.
  3. cLogistic regression, LR.