Impact of Using Uniform Attenuation Coefficients for Heterogeneously Dense Breasts in a Dedicated Breast PET/X-ray Scanner

We investigated PET image quantification when using a uniform attenuation coefficient ( $\mu $ ) for attenuation correction (AC) of anthropomorphic density phantoms derived from high-resolution breast CT scans. A breast PET system was modeled with perfect data corrections except for AC. Using uniform $\mu $ for AC resulted in quantitative errors roughly proportional to the difference between $\mu $ used in AC ( $\mu _{\mathrm{ AC}}$ ) and local $\mu $ , yielding approximately ± 5% bias, corresponding to the variation of $\mu $ for 511-keV photons in breast tissue. Global bias was lowest when uniform $\mu _{\mathrm{ AC}}$ was equal to the phantom mean $\mu $ ( $\mu _{\mathrm{ mean}}$ ). Local bias in 10-mm spheres increased as the sphere $\mu $ deviated from $\mu _{\mathrm{ mean}}$ , but remained only 2%–3% when the $\mu _{\mathrm{ sphere}}$ was 6.5% higher than $\mu _{\mathrm{ mean}}$ . Bias varied linearly with and was roughly proportional to local $\mu $ mismatch. Minimizing local bias, e.g., in a small sphere, required the use of a uniform $\mu $ value between the local $\mu $ and the $\mu _{\mathrm{ mean}}$ . Thus, biases from using uniform- $\mu $ AC are low when local $\mu _{\mathrm{ sphere}}$ is close to $\mu _{\mathrm{ mean}}$ . As the $\mu _{\mathrm{ sphere}}$ increasingly differs from the phantom $\mu _{\mathrm{ mean}}$ , bias increases, and the optimal uniform $\mu $ is less predictable, having a value between $\mu _{\mathrm{ sphere}}$ and the phantom $\mu _{\mathrm{ mean}}$ .