Room: Karl Dean Ballroom B1
Purpose: Cone-beam computed tomography (CBCT) is one of the most widely used image guidance tools in image-guided radiation therapy. However, excessive radiation exposure of patients is a concern in clinical practice. Scatter signal within the projection is another challenge degrading image quality of CBCT image. To solve the above two problems, we propose a method to perform random-ray undersampling to simultaneously reduce dose and estimate and remove scatter.
Methods: We propose a blocker with a random blocking pattern. It is constructed by sandwiching randomly placed tungsten ball-bearings between two plastic plates. The blcoker is inserted between the X-ray source and the patient. This blocker rotates during CBCT data acquisition under the control of a stepper motor to yield randomly blocked projections with varying blocking patterns in each projection. The unblocked region in the projection is utilized for CBCT reconstruction. The random projection data is favorable for compressed sensing based reconstruction. Considering that the scatter signal varies smoothly along the angular direction and in each view, the blocked region in the current view and adjacent views are used to measure the scatter signal and estimate the scatter fluence of the current view.
Results: A set of proof-of-principle simulation studies including both phantom and clinical CT data were conducted. The results showed the dose can be reduced to approximately 50% while mean relative error of scatter is within 2%. With the blocking ratio increases from 10% to 50%, structural similarity index (SSIM) gradually decreases to 96%. When the blocking ratio is further increased to 80%, SSIM is reduced to 87%.
Conclusion: The proposed random beam blocker method can potentially reduce imaging dose by approximately 50% while maintaining image quality
Funding Support, Disclosures, and Conflict of Interest: National Natural Science Foundation of China (81301940 and 81428019), National Key Research and Development Program (2016YFA0202003), Guangdong Natural Science Foundation of China (2016A030310388 and 2017A030313692), and Southern Medical University Startup fund (LX2016N003)