Gradient Step Plug-and-Play Model for Dental Cone-Beam CT Reconstruction

2026-05-27

AI Summary

This work addresses the degradation of image quality in dental cone-beam computed tomography (CBCT) caused by photon noise and proposes a novel reconstruction method that integrates data-driven priors within a plug-and-play optimization framework. Formulated as an inverse problem, the approach employs a denoising network trained on simulated noisy fan-beam projection data to approximate a proximal gradient step, which is then embedded as a learned prior into a plug-and-play gradient-based algorithm for high-quality reconstruction. Experimental results demonstrate that the proposed method significantly improves denoising performance on synthetic data and achieves superior reconstruction quality and strong generalization capability on real-world dental CBCT images.

Abstract

The goal of this work is to reduce the effect of photon noise in dental cone-beam CT reconstruction. We consider an inverse problem formulation and develop a databased prior. To this end, we simulate fan-beam acquisitions and add photon noise to the projection data. The prior is obtained by training a gradient-step denoiser using reconstructed simulated acquisitions. The trained model is integrated into a plug-and-play gradient-step algorithm to reconstruct images from simulated projections. Experiments on synthetic data demonstrate the denoising capabilities of the trained model, while qualitative evaluations on real images showcase the algorithm's performance and generalization ability.

Problem

• photon noise
• dental cone-beam CT
• image reconstruction
• denoising
• inverse problem

Innovation

• Plug-and-Play
• gradient-step denoiser
• dental cone-beam CT
• photon noise reduction
• data-driven prior