Development of digital anthropomorphic breast simulators for 3D printing

Abstract

Objective: To provide STL files for 3D printing of enhanced anthropomorphic breast simulators, including various tumour sizes, to improve mammography quality and breast cancer diagnosis accuracy. Methods: This study involved a bibliographical review encompassing breast classifications, focusing on the most common and challenging in mammography. Breast tissue density data was collected to select similar 3D printing materials. Manipulation of breast models from a repository was carried out in the 3D Builder software with defined skin and adipose tissue thicknesses. Tumors were sized in eleven different dimensions. Results: The importance of selecting dense and high glandular density breasts, which represent a significant portion of the population at high risk of breast cancer, was highlighted. Literature data on breast tissue densities and the choice of suitable 3D printing materials were crucial in creating realistic breast simulators, resulting in models of dense and fatty breasts as well as tumours of different sizes. The final files are available on the OneDrive platform for public access, providing a simulator with densities similar to real ones, contributing to the assessment of mammography image quality and enabling future tests with artificial intelligence in the detection of minimal tumours. Conclusion: The availability of anthropomorphic breast models in STL format for 3D printing offers a valuable tool in the field of radiology, enabling more objective assessments of image quality in mammography. These simulators allow for more rigorous testing, potentially enhancing diagnostic accuracy. Furthermore, in the future, they can be used to assess the limits of microcalcification detection by artificial intelligence, contributing to advancements in breast cancer detection.