Researchers at Edith Cowan University (ECU) and the University of Manitoba have created an automated program capable of detecting cardiovascular issues and fall risks from routine bone density scans.
This advancement could greatly enhance the early detection of serious health concerns before they escalate into life-threatening situations.
Developed by ECU’s Dr. Cassandra Smith and Dr. Marc Sim, the algorithm analyzes vertebral fracture assessment (VFA) images obtained during standard bone density tests, commonly used in osteoporosis treatment.
By evaluating abdominal aortic calcification (AAC) visible in these scans, the program can swiftly identify patients at risk for heart attacks, strokes, and falls.
An impressive aspect of this algorithm is its efficiency: while a skilled human operator may need five to six minutes to assess an AAC score from a single scan, the machine learning program can evaluate thousands of images in under a minute.
This level of efficiency could offer significant advantages for healthcare systems aiming to screen large populations for hidden health risks.
The necessity for such screening is clear. In the study, Dr. Smith revealed that 58% of older adults undergoing routine bone density scans presented with moderate to high levels of AAC.
Alarmingly, one in four of these patients was unaware of their heightened risk.
“Women are often under-screened and under-treated for cardiovascular conditions,” Dr. Smith remarked. “Our findings indicate that we can leverage common, low-radiation bone density machines to identify women at high risk for cardiovascular disease, enabling them to pursue necessary treatments.”
Moreover, the algorithm’s predictive capabilities extend beyond heart health. Dr. Sim’s research showed that patients with moderate to high AAC scores were also more likely to experience fall-related hospitalizations and fractures than those with low scores.
“A higher level of arterial calcification correlates with an increased risk of falls and fractures,” Dr. Sim noted. While traditional factors such as prior falls and low bone density are well recognized, vascular health is often overlooked.
“Our analysis revealed that AAC significantly contributes to fall risks, even more so than other clinically recognized risk factors.”
Like any emerging technology, there are hurdles to address before AI-assisted screening can become routine practice.
Primarily, the algorithm must be validated across larger and more diverse patient populations, and it must be effectively integrated into existing clinical workflows.
If these challenges can be overcome, a simple bone scan—something millions of older adults undergo regularly—could transform into an early warning mechanism for some of the most prevalent and severe health issues we face.
