Diffusion-based Generative Image Outpainting for Recovery of FOV-Truncated CT Images
Michelle Espranita Liman, Daniel Rueckert, Florian J. Fintelmann, Philip Müller
Abstract
Field-of-view (FOV) recovery of truncated chest CT scans is crucial for accurate body composition analysis, which involves quantifying skeletal muscle and subcutaneous adipose tissue (SAT) on CT slices. This, in turn, enables disease prognostication. Here, we present a method for recovering truncated CT slices using generative image outpainting. We train a diffusion model and apply it to truncated CT slices generated by simulating a small FOV. Our model reliably recovers the truncated anatomy and outperforms the previous state-of-the-art despite being trained on 87% less data.
Related Papers
Are Vision Foundation Models Ready for Out-of-the-Box Medical Image Registration?2025-07-15PSAT: Pediatric Segmentation Approaches via Adult Augmentations and Transfer Learning2025-07-08SV-DRR: High-Fidelity Novel View X-Ray Synthesis Using Diffusion Model2025-07-07Causal-SAM-LLM: Large Language Models as Causal Reasoners for Robust Medical Segmentation2025-07-04Pose-Star: Anatomy-Aware Editing for Open-World Fashion Images2025-07-04X-SiT: Inherently Interpretable Surface Vision Transformers for Dementia Diagnosis2025-06-25Volumetric segmentation of muscle compartments using in vivo imaging and architectural validation in human finger flexors2025-06-25Learning from Anatomy: Supervised Anatomical Pretraining (SAP) for Improved Metastatic Bone Disease Segmentation in Whole-Body MRI2025-06-24