Daijiworld Media Network - Winnipeg

Winnipeg, Apr 30: In a breakthrough for preventive healthcare, researchers from Australia and Canada have developed an advanced machine learning algorithm that can rapidly detect heart disease and fracture risks using standard bone density scans.

The technology, created by Edith Cowan University (ECU) and the University of Manitoba, aims to enhance early detection of serious health issues during routine osteoporosis screenings, potentially benefitting millions of older adults, according to Xinhua news agency.

The automated system analyses vertebral fracture assessment (VFA) images to identify abdominal aortic calcification (AAC) a silent but potent predictor of cardiovascular events such as heart attacks and strokes, as well as falls and fractures.

Traditionally, assessing AAC required five to six minutes per scan by trained experts. The new algorithm cuts that down to under a minute per image, enabling faster and larger-scale screening.

“About 58 per cent of older women undergoing bone scans showed moderate to high levels of AAC often without knowing they were at risk,” said ECU research fellow Cassandra Smith. She noted that cardiovascular disease in women remains under-screened and under-treated.

“Without specific screening for AAC, these signs would go unnoticed. This algorithm helps ensure earlier and more accurate diagnoses,” she added.

Further findings by ECU’s Marc Sim revealed that AAC is also a powerful predictor of falls and fractures, surpassing traditional markers like bone density and previous fall history.

“The higher the calcification in your arteries, the higher the risk of falls,” Sim explained. “Vascular health is often overlooked in fall risk assessments, and this technology brings it to the forefront.”

He added that integrating the algorithm into routine bone scans could significantly enhance how clinicians evaluate a patient’s overall fracture and cardiovascular risk marking a key step forward in preventive medicine.