VesselFusion: Diffusion Models for Vessel Centerline Extraction from 3D CT Images
This work provides an improved method for medical image analysis, specifically for researchers and clinicians needing accurate and natural vessel centerline extraction from 3D CT images, offering an incremental improvement over existing deterministic models.
This paper addresses the challenge of extracting vessel centerlines from 3D CT images, a task crucial for reducing annotation effort in building vessel structure estimation models. The authors propose VesselFusion, a diffusion model that employs a coarse-to-fine representation and voting-based aggregation, achieving higher extraction accuracy and more natural results compared to conventional deterministic approaches on a public dataset.
Vessel centerline extraction from 3D CT images is an important task because it reduces annotation effort to build a model that estimates a vessel structure. It is challenging to estimate natural vessel structures since conventional approaches are deterministic models, which cannot capture a complex human structure. In this study, we propose VesselFusion, which is a diffusion model to extract the vessel centerline from 3D CT image. The proposed method uses a coarse-to-fine representation of the centerline and a voting-based aggregation for a natural and stable extraction. VesselFusion was evaluated on a publicly available CT image dataset and achieved higher extraction accuracy and a more natural result than conventional approaches.