Amgen – Crystal Bone Algorithm May Predict Osteoporosis Fracture Risk

Lead developer, Yasmeen Almog, a data scientist on Amgen’s Digital Health and Innovation Team discusses Crystal Bone, a new digital health algorithm under development that may help doctors predict fracture risk in osteoporosis patients within a one to two-year time frame. 

Yasmeen Almog is a data scientist on Amgen’s Digital Health and Innovation Team. Since joining Amgen in 2018, she has worked on a variety of projects that address diverse challenges at Amgen, ranging from commercial to R&D applications. Most prominently, Yasmeen led the ideation and development of Crystal Bone, a deep learning algorithm that utilizes Electronic Health Record data to predict short-term risk of fracture due to osteoporosis, which has become a focus of Amgen’s broader AI strategy. This work has been recognized by the scientific community, resulting in publication and presentations with the Endocrine Society, the American Society of Bone and Mineral Research, and most recently, the Journal of Medical Internet Research. Her work has also included an international rotation in Reykjavik, Iceland, where she helped to bring next-generation artificial intelligence capabilities to the multi-omics research at deCODE Genetics. Prior to Amgen, Yasmeen earned a Bachelor’s of Science and Engineering (B.S.E.) in Computer Science from Princeton University. She was first introduced to data science through her studies, particularly her senior thesis research, in which she developed an algorithm that utilized computer vision and machine learning in an attempt to remove human bias from the judging in springboard diving, a sport she competed in throughout her undergraduate experience.