{"ID":2881114,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.12900","arxiv_id":"2508.12900","title":"CTFlow: Video-Inspired Latent Flow Matching for 3D CT Synthesis","abstract":"Generative modelling of entire CT volumes conditioned on clinical reports has the potential to accelerate research through data augmentation, privacy-preserving synthesis and reducing regulator-constraints on patient data while preserving diagnostic signals. With the recent release of CT-RATE, a large-scale collection of 3D CT volumes paired with their respective clinical reports, training large text-conditioned CT volume generation models has become achievable. In this work, we introduce CTFlow, a 0.5B latent flow matching transformer model, conditioned on clinical reports. We leverage the A-VAE from FLUX to define our latent space, and rely on the CT-Clip text encoder to encode the clinical reports. To generate consistent whole CT volumes while keeping the memory constraints tractable, we rely on a custom autoregressive approach, where the model predicts the first sequence of slices of the volume from text-only, and then relies on the previously generated sequence of slices and the text, to predict the following sequence. We evaluate our results against state-of-the-art generative CT model, and demonstrate the superiority of our approach in terms of temporal coherence, image diversity and text-image alignment, with FID, FVD, IS scores and CLIP score.","short_abstract":"Generative modelling of entire CT volumes conditioned on clinical reports has the potential to accelerate research through data augmentation, privacy-preserving synthesis and reducing regulator-constraints on patient data while preserving diagnostic signals. With the recent release of CT-RATE, a large-scale collection...","url_abs":"https://arxiv.org/abs/2508.12900","url_pdf":"https://arxiv.org/pdf/2508.12900v1","authors":"[\"Jiayi Wang\",\"Hadrien Reynaud\",\"Franciskus Xaverius Erick\",\"Bernhard Kainz\"]","published":"2025-08-18T12:58:21Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[\"Transformer\",\"Variational Autoencoder\"]","has_code":false}