Last week, Rudolfs Latisenko, data scientist at Thirona,  presented our deep learning approach for reducing dose-related noise on low dose CT at the COPDGene 2019 fall meeting in Boston. The developed model estimates conventional dose CT from reduced dose acquisitions using Cycle-consistent Generative Adversarial Networks (CycleGAN) and was developed in collaboration with Stephen Humphries and David Lynch from National Jewish Health. Rudolfs did an excellent job explaining the impact of our work to the community and showed that our method substantially reduces the dose-introduced variation in CT quantified emphysema within COPDGene.