Purpose: This study aimed to develop a 3D-printed lithophane breast anthropomorphic phantom for optimizing the automatic exposure control (AEC) in a digital mammography system, thereby reducing radiation dose while maintaining high image quality.
Methods: Craniocaudal breast radiograhic images from 72 patients, categorized as high-density and low-density by radiologists, were used to design the phantom. A digital lithophane technology was employed to create an anatomic breast plate, fabricated using a digital light processing 3D printer with resin. Polymenthylmethacrylate (PMMA) support thickness was adjusted incrementally until the exposure index and deviation index values approximated those of the American College of Radiology phantom. Phantom images were acquired across five AEC density levels (−6, −3, 0, 3, 6), and the optimal dose was determined as the lowest autoexposure mAs value with superior image quality. Two radiologists scored image quality on a 7-point Likert scale to identify the best configurations.
Results: The optimal PMMA support thicknesses were determined as 3 cm for high-density and 4 cm for low-density breasts. The optimized AEC condition corresponded to the lowest density level (−6) with the least mAs value, maintaining excellent image quality. The use of the phantom resulted in a reduction of automatic exposure tube current by 39.4%–43.4% while producing images comparable to human breast radiographic images.
Conclusions: The developed 3D-printed lithophane breast anthropomorphic phantom effectively optimized AEC settings, reducing radiation dose and maintaining high-quality breast radiographic images. This study has the potential to enhance safety and diagnostic efficacy in digital mammography.

KeywordsBreast anthropomorphic phantom, Digital lithophane technology, Automatic exposure control, Dose optimization, 3D printing technology