This study presents a deep learning approach to parametric response mapping (PRM) for small airway disease (SAD) screening using inspiratory chest CT scans. Traditionally, PRM requires both inspiratory and expiratory scans. The study involved 537 participants with normal spirometry and a history of smoking or secondhand smoke exposure, divided into training, tuning, and test sets. A cascaded generative adversarial network generated expiratory CT from inspiratory CT, followed by a UNet-like network predicting PRM using real inspiratory and generated expiratory CT. The performance was evaluated using SSIM, RMSE, dice coefficients, and Pearson correlation. Aview software (Coreline Soft, Seoul, Korea) was used to validate the accuracy of the ground truth PRM. The method generated high-quality expiratory CT (SSIM 0.86, RMSE 80.13 HU). Predicted PRM dice coefficients for normal lung, emphysema, and fSAD were 0.85, 0.63, and 0.51, respectively. The deep learning method effectively stratified SAD, demonstrating its potential for evaluating SAD using only inspiratory CT scans.