Author: Ahad Ollah Ezzati, Yile Fang, Xiaoyu Hu, Xun Jia, Kai Yang, Yuncheng Zhong 👨🔬
Affiliation: Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University, Massachusetts General Hospital, Johns Hopkins University 🌍
Purpose: With breast cancer being one of the most prevalent cancers, regular screening is an effective strategy to mitigate the risk of malignancy. However, conventional energy-integrating detector (EID) based CT scanner cannot be adapted due to many limitations, such as its large bore size, low spatial resolution, and high noise under acceptable dose level. Photon-counting detectors (PCD) have emerged as a superior solution over the EIDs allowing the detection of suspicious microcalcifications. This study proposes a compact PCD-based static cone-beam CT (PCD-sCBCT) system for dedicated breast imaging.
Methods: The proposed imaging system consisted of 51 PCD panels (2.5 40 cm) connected side-to-side with their longer side to form a 51-sided polygon from a top view. 306 x-ray sources were installed on the top side of each PCD panel. The source-to-axis distance was 20.3 cm, and the source-to-detector distance was 40.6 cm. A CT scan of a digital anthropomorphic breast phantom was simulated with the in-house developed Monte-Carlo code under the configuration of the proposed system, generating 306 projections over 2 angular range. Images were reconstructed via a modified filtered backprojection algorithm. Two important issues were investigated with our simulation platform: scatter control via narrowly collimated cone-beam acquisition scheme, and the source motion effect via simulation of a ball bearing digital phantom.
Results: The breast phantom image demonstrated the imaging capabilities of the compact PCD-sCBCT system. The scatter-to-primary ratio (SPR) was reduced by approximately 90% through the implementation of the collimated x-ray source scanning protocol. The full width at half maximum of the ball bearing was reduced by 45.6% indicating that the static design effectively eliminated source motion artifacts.
Conclusion: Preliminary results indicate the feasibility of the proposed PCD-sCBCT system for dedicated breast imaging. Ongoing investigations involve optimizations of system design, data acquisition protocols and image quality improvement.