Using Machine Learning to Predict Progression in the Gastric Precancerous Process in a Population from a Developing Country Who Underwent a Gastroscopy for Dyspeptic Symptoms

Table 3

Results of linear regression for the predictors identified by best subset linear regression with the leave-one-out method.

Outcome	Predictors	Regression coefficients (95% confidence intervals)
Overall progression (general model)	Complete intestinal metaplasia	0.534 (0.425, 0.644)
	Incomplete intestinal metaplasia	0.316 (0.187, 0.444)
	Histological diagnosis at baseline less advanced than atrophic gastritis	-0.313 (-0.368, -0.258)
	Depth of corpus inflammation at baseline	-0.152 (-0.265, -0.039)
	Average density of polymorphonuclear cells in the antrum at baseline	0.037 (-0.029, 0.103)
	Alcohol intake at baseline	-0.189 (-0.201, 0.022)
Overall progression (location-specific model)	Complete intestinal metaplasia at baseline	0.492 (0.382, 0.602)
	Incomplete intestinal metaplasia at baseline	0.345 (0.223, 0.467)
	Histological diagnosis at baseline less advanced than atrophic gastritis	-0.296 (-0.351, -0.241)
	Depth of corpus inflammation at baseline	-0.150 (-0.264, -0.037)
	Density of H. pylori infection in the corpus and the antrum at baseline	0.122 (0.030, 0.214)
	Intake of fried fava beans per week at baseline	0.064 (0.0004, 0.128)