Dental CT scans extend field of view with AI and physics

Field-of-View Extension in Dental Cone-Beam CT via Implicit Neural Representations and Diffusion Model-Based Refinement

Computer Vision and Pattern RecognitionArtificial Intelligence

Summary

Cone-beam CT scanners used in dentistry often capture only a small part of the mouth due to a limited field of view. The authors developed a three-step method that first fills in missing scan data using a special type of neural network, then reconstructs the extended image with physics-based calculations, and finally enhances the image using an AI model called a diffusion model. Their approach helps produce clearer and more complete images of a patient’s teeth and jaws. They also made their code available for others to use.

What this means in practice

  • For dental imaging teams: Generate more complete dental CT images despite limited scan coverage to improve diagnosis and treatment planning.
  • For medical image software developers: Incorporate a pipeline that combines neural data completion, iterative reconstruction, and diffusion-based denoising to improve truncated CT volumes.

Authors

Susanne Schaub, Florentin Bieder, Matheus L. Oliveira, Yulan Wang, Buyanbileg Sodnom-ish, Dorothea Dagassan-Berndt, Michael M. Bornstein, Philippe C. Cattin

Abstract

Dental cone-beam computed tomography (CBCT) systems often employ detector configurations that provide a truncated field of view (FOV) that only captures a small part of the patient's anatomy. In this work, we aim to reconstruct an extended FOV using projections of truncated FOV scans. To this end, we propose a three-stage framework that consists of (1) an implicit neural representation (INR) for estimating missing parts of the truncated projection data, (2) an iterative reconstruction for generating a secondary volumetric image with improved anatomical consistency and (3) a fast diffusion model for image enhancement. The proposed approach combines the strengths of continuous representations, physics-based reconstruction and generative refinement within a unified pipeline for truncated CBCT imaging. Experimental results demonstrate that the method effectively reduces truncation artifacts, improves the reconstruction of structures extending beyond the original FOV and produces images with enhanced quality. Our code is publicly available at https://github.com/SusanneSchaub/CBCT-FOV-Extension.