Suborgan breast dosimetry for breast nuclear medicine imaging using anthropomorphic software breast phantoms

Abstract Breast nuclear medicine imaging has gradually become a supplemental method for detection of mammographically occult cancer in dense breasts. However, mammary glands and ducts are highly sensitive to radiation; the absorbed dose to glandular tissue may increase the risk of secondary breast cancer. Therefore, accurate suborgan breast dosimetry for breast nuclear medicine imaging represents an important issue. In this study, a series of anthropomorphic software breast phantoms (ASBPs) with various volumes and glandularities were built. The realism of the ASBPs was assessed by using the two-dimensional projection and fractal dimension (FD) analysis. Breast self S values and suborgan S values for 18F and 99mTc were simulated using the Monte Carlo technique. The projected images of the ASBPs were similar to clinical mammograms in tissue distribution and image contrast. The mean FD value for the ASBPs was 2.23. The maximum differences between the breast self S values of ASBPs and OLINDA/EXM for 18F and 99mTc were −14.4% and −16.7%, respectively. For the breast suborgan S values, the maximum ratios between the S value from gland to gland S(g←g) and the S value from breast to breast S(b←b) were 4.84 for 18F and 3.60 for 99mTc. The mean ratios between the S value from gland to duct S(d←g) and that from gland to gland S(g←g) were 0.296 for 18F and 0.356 for 99mTc. The suborgan breast dosimetry based on the ASBPs can be used for internal dose evaluation and carcinogenic risk assessment of mammary glands and ducts in breast nuclear medicine imaging.