Enhancing Synthetic CT from CBCT via Multimodal Fusion: A Study on the Impact of CBCT Quality and Alignment
This work addresses image quality issues in real-time intraoperative imaging for medical applications, but it is incremental as it builds on existing synthetic CT generation methods.
The study tackled the problem of artifacts in Cone-Beam CT (CBCT) by enhancing synthetic CT generation through multimodal learning that integrates intraoperative CBCT with preoperative CT, resulting in consistent outperformance over unimodal baselines, with significant gains in well-aligned, low-quality cases.
Cone-Beam Computed Tomography (CBCT) is widely used for real-time intraoperative imaging due to its low radiation dose and high acquisition speed. However, despite its high resolution, CBCT suffers from significant artifacts and thereby lower visual quality, compared to conventional Computed Tomography (CT). A recent approach to mitigate these artifacts is synthetic CT (sCT) generation, translating CBCT volumes into the CT domain. In this work, we enhance sCT generation through multimodal learning, integrating intraoperative CBCT with preoperative CT. Beyond validation on two real-world datasets, we use a versatile synthetic dataset, to analyze how CBCT-CT alignment and CBCT quality affect sCT quality. The results demonstrate that multimodal sCT consistently outperform unimodal baselines, with the most significant gains observed in well-aligned, low-quality CBCT-CT cases. Finally, we demonstrate that these findings are highly reproducible in real-world clinical datasets.