Abstract While most of the anatomy of the human body is well delineated and understood, the distribution of adult female breast ducts remains largely understudied despite the fact that breast cancer, is thought to start in the lining of a breast ductal lobular unit. These anatomical features develop from stem cells located behind the nipple under the influence of the hormones of puberty. The exact patterns and number of ducts in the normal female breast has received limited attention since the work of Astley Cooper in 1840 who admitted that he had artfully displayed the ducts in two dimensions for his book. Since then, Going et al has used computerized reconstruction to map one breast and investigators in Australia have used hand-held ultrasound on lactating women to gain some insight into the nipple anatomy. The descriptions of the number of ducts exiting the breast at the nipple can be divided into those who have cannulated ducts at the nipple (6-10) and those who enumerated the number of ducts seen on transection of the nipple. (15-20). Local anesthesia can be used as a nipple block allowing duct cannulation in women. Breast cancer starts in the lining of a milk duct which opens the way for the application of new diagnostic techniques such as liquid biopsy. The limitation of this approach, however, is the absence of an imaging modality to direct sampling. In collaboration with QT Imaging, images were obtained using non-invasive 3D Transmission Ultrasound (TU) imaging of whole in vivo breasts of three patients, a postmenopausal 64 year old woman, a woman with DCIS and one with an invasive cancer. TU is a novel FDA cleared imaging device utilizing 3D ultrasound data and modelling to yield speckle free sub-mm resolution quantitative estimates of tissue characteristics. The women were lying prone with their breast immersed in a water bath with reconstruction of the projection and reflected ultrasound data resulting in co-registered 3D reflection, speed-of-sound, and attenuation images. Machine learning at a voxel level was used to quantitively differentiate and segment breast tissues types based on reconstructed voxel tissue characteristics. These tissue types were then 3D printed into anatomical models representing a normal postmenopausal breast, one with DCIS and one with invasive cancer. The images demonstrated the distribution of ductal systems in two concentric groupings unrelated to quadrants. The DCIS and the invasive ductal cancer were limited to a single ductal system each. This new non-invasive technology enables precise localization of intraductal and invasive lesions. It opens up the potential for precise anatomic intraductal liquid biopsy for diagnosis as well as intraductal therapy. In addition, it has ramifications for better directing surgical removal of intraductal and invasive lesions. Citation Format: Susan M Love, James Klock, James Wiskin. Imaging of the distribution of human breast ducts [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS3-15.