Diffusion-based Generative Image Outpainting for Recovery of FOV-Truncated CT Images
This addresses the need for accurate disease prognostication through body composition analysis in medical imaging, but it is incremental as it builds on existing generative methods for a specific domain.
The authors tackled the problem of recovering truncated chest CT scans for body composition analysis by using a diffusion-based generative image outpainting method, which reliably recovers anatomy and outperforms the previous state-of-the-art while being trained on 87% less data.
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.