Researchers in Limerick have developed a new way of using routine bone density scans (DEXA) to predict the age at which someone is likely to develop osteoporosis and when treatment could help them recover from it. Osteoporosis is the most common bone condition and is characterised by reduced bone strength leading to increased fracture (4 people per hour in Ireland).
The study, just published in the leading Medical Informatics & Decision Making journal, was carried out by researchers at University Hospital Limerick and the Research Directorate (HSE Midwest), analysed repeated DEXA scans from 200 patients. Download paper here.
Currently, DEXA scans are mainly used to diagnose osteoporosis by measuring bone density at a single point in time. The new research instead uses a series of scans to estimate a patient’s future bone health trajectory.
The researchers created mathematical models that can estimate:
- the age at which a person may enter the osteoporotic range, and
- the age at which they may leave that range following treatment or recovery.
The team says the approach should help doctors explain bone health in more intuitive, age-based terms.
“Patients often want to know not just what their bone density is at a moment in time, but where it is heading,” said Dr Azrin Muslim, Research Fellow. “This framework aims to translate bone scans into understandable timelines using language familiar to patients.”
Unlike many artificial intelligence systems used in healthcare, the researchers say their model is fully transparent, with every prediction traceable directly to the patient’s scan results.
The study is an early proof-of-concept and has potential for widespread clinical applicability. The authors are optimistic that this approach could enhance the management of osteoporosis worldwide and believe that the findings could support earlier identification of high-risk patients, improve communication around osteoporosis risk, and help monitor recovery during treatment.
