Abstract
Breast cancer is one of the main reasons that lead to mortality, particularly in females. One of the factors that help in breast cancer detection is the existence of breast phantom with high similarity to the real breast associated with the utilized modality. Currently, the existing breast phantoms are limited in terms of anatomy and breast ionization reactions properties. Thus, the aim of this paper is to design a heterogenous patient-based breast phantom that mimics the real breast tissues properties concerning ionizing imaging modalities. The phantom includes skin tissue, fibroglandular tissue, adipose tissue, pectoral muscles, and insertion of malignant lesion. The process started with real breast MR images from King Fahd Specialist Hospital with BI-RADS I tissue segmentation for molds creation. 3D Slicer software was used for segmentation, and the Meshmixer was used to refine the segmented tissue. The molds are printed using PRUSA 3D printers. The tissue-mimicking materials were tailored in terms of their elemental composition weight fractions and three ionization properties. These are the mass attenuation coefficient (MAC), electron density (n e ) and effective atomic number (Z eff ). These characteristics were compared with the corresponding properties of the real breast ICRU reported elemental compositions. The achieved results showed an excellent agreement between the MAC of the tissue-mimicking materials and the ICRU-based breast tissues. The percentage of error in n e and Z eff amounts to only 2.93% and 5.76%, respectively. That means the phantom can optimize the function of breast ionizing imaging modalities and lead to higher breast cancer detection sensitivity.