Estimating 3D Dental Structures using Simulated Panoramic Radiographs and Neural Ray Tracing

Panoramic radiography (Panoramic X-ray, PX) is a widely used imaging modality for dental examination. Since PX only provides 2D flattened views of the oral structure, its applicability is limited as compared to 3D Cone-beam computed tomography (CBCT). In this paper, we propose a framework to estimate CBCT-like 3D structures from real-world PX. Our framework tackles full 3D reconstruction for varying subjects (patients) where each reconstruction is based only on a single panoramic image. We create an intermediate representation called simulated PX (SimPX) from CBCT data which is based both on the Beer-Lambert law of X-ray rendering and rotational principles of PX imaging. SimPX aims at not only truthfully simulating PX, but also facilitates the reverting process back to 3D data. We propose a novel neural model based on ray tracing which exploits both global and local input features to convert SimPX to 3D output. At inference, a real PX image is translated to a SimPX-style image with semantic regularization, and the translated image is processed by generation/refinement modules to produce high-quality outputs. Experiments show that our method outperforms prior state-of-the-art in reconstruction tasks both quantitatively and qualitatively. Our method does not require any prior information such as the shape of dental arches, nor the matched PX-CBCT dataset for training, which is difficult to obtain in clinical practice.