Common fracture risk assessment tools e.g. FRAX and Garvan, confer long-term but not short-term risk estimates necessary to identify patients likely to fracture in the next 1–2 years. Furthermore, these tools utilise cross-sectional data representing a subset of all available clinical risk factors for risk prediction. Thus, these methods are generalised across patient populations and may not fully utilise patient histories in electronic health records (EHRs) that contain temporal information for thousands of unique features.
We used the Optum® EHR dataset to develop Crystal Bone, a method that applies machine learning techniques to predict fracture risk over a 1‒2 year timeframe. Crystal Bone uses context-based embedding techniques to learn an equivalent “semantic” meaning of various medical events. Similar to how language models predict the next word in a given sentence, Crystal Bone can predict that a patient’s future trajectory may contain a fracture or that the “signature” of the patient’s overall journey is similar to that of a typical fracture patient.
We applied Crystal Bone to two datasets, one enriched for fracture patients and one representative of a typical hospital system. When predicting likelihood of fracture in the next 1–2 years, Crystal Bone had an area under the receiver operating characteristic (AUROC) score ranging from 72%‒83% on a test (hold-out) dataset. These results suggest performance similar to FRAX and Garvan, which have 10-year fracture risk prediction AUROC scores of 64.4% +/– 3.7%.
In conclusion, it is possible to use each patient’s unique medical history as it changes over time to predict patients at risk for fracture in 1–2 years. Furthermore, it is theoretically possible to integrate a model like Crystal Bone directly into an EHR system, enabling “hands-off” fracture risk prediction, which could lead to improved identification of patients at very high risk for fracture.