Reconstruction of MVCT Based on Irregular Portal Images Acquired by EPID
In the realm of radiation therapy, the precise positioning of patients constitutes a critical procedure, as it serves to ensure the accurate delivery of radiation doses and minimize the potential damage inflicted upon normal tissues. Accurate positioning not only enhances treatment efficacy but also significantly reduces the likelihood of adverse effects and complications. The Electronic Portal Imaging Device (EPID) is directly integrated into the gantry of a linear accelerator. By using the same source with the treatment, EPID eliminates the mechanical deviation issues inherent in traditional Cone Beam Computed Tomography (CBCT) systems and enables true real-time dose verification. However, applying images acquired by EPID to Mega Voltage Computed Tomography (MVCT) reconstruction and positioning verification still faces significant challenges. EPID captures irregular portal images during treatment, resulting in incomplete projection data, uneven projection angles, and even projection angles missing—all of which pose difficulties for MVCT reconstruction. Additionally, compared with CBCT systems, MVCT shows significantly lower soft-tissue contrast and signal-to-noise ratio. This paper proposes a new method for reconstructing MVCT from irregular portal images acquired by EPID. First, the two-dimensional irregular portal images synchronously collected by EPID during treatment are used as the original data source. After preprocessing, the ART reconstruction algorithm is applied to the pre-processed data, with TV regularization added during the reconstruction process for optimization. Results from the simulations confirm that the proposed algorithm is feasible, satisfactory reconstruction results can be obtained when $\lambda=0.3$ and paceTV=2.0. This method requires no additional operations and does not increase the patient’s radiation dose. The reconstructed MVCT provides technical prerequisites for subsequent positioning verification.
Authors
- Zerui Chen
- Wei Zhan
- Qian Cao
- Junhai Wen
Institutions
- Beijing Institute of Technology (CN)
Publication Details
- Published
- 2025-06-20
- DOI
- https://doi.org/10.1109/eee64897.2025.11162770
- Primary Topic
- Advanced MRI Techniques and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00