HARU-Net: Hybrid Attention Residual U-Net for Edge-Preserving Denoising in Cone-Beam Computed Tomography

OBJECTIVES: Cone-beam computed tomography (CBCT) is widely used in dental and maxillofacial imaging, but low-dose acquisition introduces strong, spatially varying noise that degrades soft-tissue visibility and obscures fine anatomical structures. Classical denoising methods struggle to suppress noise in CBCT while preserving edges. Although deep learning-based approaches offer high-fidelity restoration, their use in CBCT denoising is limited by the scarcity of high-resolution CBCT data for supervised training. This study aims to develop an efficient deep learning framework for high-quality CBCT denoising that effectively suppresses noise while preserving fine anatomical structures and maintaining computational efficiency for practical clinical deployment. APPROACH: To achieve this objective, we propose a novel Hybrid Attention Residual U-Net (HARU-Net) for high-quality denoising of CBCT data, trained on a cadaver dataset of human hemimandibles acquired using a high-resolution protocol of the 3D Accuitomo 170 (J. Morita, Kyoto, Japan) CBCT system. The novel contribution of this approach is the integration of three complementary architectural components: (i) a hybrid attention transformer block (HAB) embedded within each skip connection to selectively emphasize salient anatomical features, (ii) a residual hybrid attention transformer group (RHAG) at the bottleneck to strengthen global contextual modeling and long-range feature interactions, and (iii) residual learning convolutional blocks to facilitate deeper, more stable feature extraction throughout the network. RESULTS: HARU-Net\footnote{https://github.com/DrKay87/HARU-Net--Hybrid-Attention-Residual-U-Net-for-Denoising-in-Cone-Beam-Computed-Tomography} consistently outperforms state-of-the-art (SOTA) methods achieving the highest PSNR (37.52 dB), the second-highest SSIM (0.9557), and the lowest GMSD (0.1084). Compared with transformer-based methods, the proposed network achieves superior denoising performance while maintaining substantially lower computational complexity. CONCLUSION: The proposed HARU-Net provides an effective balance between noise suppression, anatomical structure preservation, and computational efficiency. These characteristics make it a promising and practical solution for improving image quality and supporting reliable diagnosis in low-dose CBCT imaging.