Pulmonary impairments are observed as decreased changes in lung density during respiration on dynamic chest radiographs (DCR). To facilitate pulmonary function evaluation based on DCR, the present study was conducted to correlate respiratory changes in pixel values (▵pixel values) on DCR images with those in computed tomography (CT) values (▵CT values) using four-dimensional extended cardiac‐torso (XCAT) phantoms. Twenty XCAT phantoms with forced breathing with/without chest wall motion were generated, and then, were projected using an X-ray simulator. ▵pixel values and ▵CT values on respiration were calculated on the simulated projection and CT images, respectively, to create a conversion table. The equation (regression line) of the relationship between the ▵pixel values and ΔCT values in the lung field was calculated, and statistical analysis was performed to test whether there was a significant difference in the inclination of the regression line (P < 0.05) in terms of physique, sex, and breathing manner. There was a significant difference in the inclination of the regression lines between males and females, with and without chest wall motion. We developed a conversion table from ▵pixel values to ▵CT values in the lung field. Our results demonstrated the feasibility of the quantitative evaluation of pulmonary function based on ▵pixel values on DCR. It would be helpful to provide a better understanding of the pulmonary function and conditions based on ▵pixel values while considering ▵CT values.