Figure 3
A method for visualizing changes in the morphology of teeth with age using the coefficients obtained from the trained linear regression model. β, the coefficient of the regression model trained with the latent variables and actual age data, represents the correlation between the latent variables and age. Thus, when adding or subtracting the coefficient β to or from the original latent variables generated from the reference image, the values of the latent variables that have a strong correlation with age tend to change greatly, while the values of those that have a weak correlation with age change less significantly. Therefore, the values of the latent z variable change according to the degree of correlation with age, and this relationship can be visually demonstrated using the decoder.
