MeshBrush: Painting the Anatomical Mesh with Neural Stylization for Endoscopy
This work addresses the sim-to-real gap in medical endoscopy, specifically for ureteroscopy, with potential transferability to other endoscopic procedures, though it is incremental as it builds on existing image-to-image translation methods.
The paper tackles the problem of generating temporally consistent synthetic endoscopic videos from pre-operative scans by introducing MeshBrush, a neural mesh stylization method that uses learned per-vertex textures to produce high-fidelity outputs, demonstrating its effectiveness for tasks like training networks and preoperative planning.
Style transfer is a promising approach to close the sim-to-real gap in medical endoscopy. Rendering synthetic endoscopic videos by traversing pre-operative scans (such as MRI or CT) can generate structurally accurate simulations as well as ground truth camera poses and depth maps. Although image-to-image (I2I) translation models such as CycleGAN can imitate realistic endoscopic images from these simulations, they are unsuitable for video-to-video synthesis due to the lack of temporal consistency, resulting in artifacts between frames. We propose MeshBrush, a neural mesh stylization method to synthesize temporally consistent videos with differentiable rendering. MeshBrush uses the underlying geometry of patient imaging data while leveraging existing I2I methods. With learned per-vertex textures, the stylized mesh guarantees consistency while producing high-fidelity outputs. We demonstrate that mesh stylization is a promising approach for creating realistic simulations for downstream tasks such as training networks and preoperative planning. Although our method is tested and designed for ureteroscopy, its components are transferable to general endoscopic and laparoscopic procedures. The code will be made public on GitHub.