SU‐E‐I‐17: Evaluation of Cone Beam Breast CT (CBCT) System: Detection of Randomly Distributed Micro‐Calcification and the Effect of Beam Hardening Filter

Abstract

Purpose: Detectability of micro‐calcification is very important for breast imaging. The purpose of this study is to evaluate the breast CBCT system through its detectability of randomly distributed micro‐calcifications under different mean glandular dose (MGD) and with or without different thickness copper filters. Methods: Calcium carbonate grains of various sizes (Three size groups: 224–250μm, 250–280μm, 280–300μm) were used to simulate calcifications. A breast shaped plastic container filled with paraffin was used to simulate a fatty breast; the top half of the phantom was cylindrical with 12.5cm diameter and the total height of the phantom is 14.5cm. The phantom was imaged with a CBCT system in our lab. The system consists of a conventional tungsten target x‐ray tube (G‐1592) and an a‐Si:H/CsI flat panel detector (Paxscan 4030CB). The source to image detector and source to iso‐center distances were 152 and 100 cm, respectively, resulting in a magnification factor of 1.52 at the iso‐center. All the scans were conducted at 80 kVp. The experiment was conducted under MGD equal to 2.5, 4, 6, 8 mGy. With each MGD setting, the scan was conducted without filter, with 0.002″ copper filter or with 0.004″ copper filter. Feldkamp algorithm with a ramp filter was used for image reconstruction of all scans. Results: Our CBCT system is capable of imaging calcifications as small as 224–250μm with 6mGy MGD or 250–280μm with 4mGy MGD if set 50% visibility as threshold. To all three size groups, filtering improves visibility of micro‐calcification when the dose is between invisible and high visible settings. Conclusions: Filtering improves detectability of micro‐calcification for low dose image.This work was supported in part by grants CA104759, CA124585 and CA13852 from NIH‐NCI, a grant EB00117 from NIH‐NIBIB, and a subcontract from NIST‐ATP.