AsianScientist (Jul. 09, 2026) – Osteoporosis is usually known as a “silent illness” as a result of bones weaken regularly over a few years with out inflicting signs – till a fracture happens. Detecting the situation early is essential, however present medical screening suggestions primarily deal with older girls and different high-risk teams, which leaves some males, and youthful adults with undiagnosed bone loss.
Osteoporosis impacts over 200 million people globally and contributes to greater than 8.9 million fractures yearly. As populations age, particularly in Asia the place demographic transitions are accelerating, the incidence of osteoporosis-related fractures is projected to rise dramatically, with over 50 per cent of worldwide hip fractures predicted to happen in Asia by 2050.
Now, a brand new research by researchers from Nationwide Taiwan College and St. Paul’s Hospital has proven that synthetic intelligence (AI) can extract important bone-health info from routine chest X-rays.
Printed in npj Digital Drugs, the research demonstrates how AI can extract details about bone well being from chest X rays which are already extensively carried out throughout routine well being check-ups throughout Asia. The researchers state that this method might assist determine individuals who wouldn’t usually qualify for osteoporosis screening however who could nonetheless be vulnerable to fractures.
“This research demonstrates how synthetic intelligence can remodel current healthcare workflows into scalable preventive-health methods whereas supporting extra equitable entry to osteoporosis screening,” mentioned Ray-E Chang, who’s a co-corresponding creator of the research, and a professor on the Institute of Well being Coverage and Administration in Nationwide Taiwan College, Taipei.
Present osteoporosis screening suggestions focus totally on postmenopausal girls, older adults and other people with established medical threat elements. In consequence, youthful adults, males and people with a wholesome physique weight usually stay outdoors routine screening pathways. Based on the research, this method could overlook a considerable proportion of individuals with lowered bone density.
Greater than half of the contributors who had been confirmed to have irregular bone density had a traditional physique mass index (BMI), highlighting a serious blind spot in standard risk-based screening.
As a substitute of changing current diagnostic strategies, the AI system is designed to determine people who could profit from additional evaluation utilizing dual-energy X-ray absorptiometry (DXA), the present gold commonplace for measuring bone mineral density.
“Below Taiwan’s Nationwide Well being Insurance coverage system, we regularly depend on strict guideline-based standards to resolve who qualifies for DXA testing,” mentioned Shu-Han Chen, MD, first creator of the research, a household drugs doctor, and chief of the Well being Administration Heart at St. Paul’s Hospital in Taoyuan.
“Our findings counsel that AI-assisted chest X-ray evaluation might assist determine people who could in any other case be neglected and who could profit from confirmatory DXA testing,” Shu-Han added.
One of many key benefits of the method is that it could repurpose imaging already being collected throughout routine medical care. As a result of no extra scans are required, opportunistic screening might be launched with little further value or inconvenience for sufferers. By analysing current photographs, healthcare suppliers could possibly determine early bone loss earlier than fractures happen, enabling well timed remedy and life-style interventions.
—
Supply: t. Paul’s Hospital, Taiwan ; Picture: The Yuri Arcurs Assortment/magnific
You will discover the research at: Advancing diagnostic fairness via synthetic intelligence chest radiograph screening for osteoporosis in Asian populations
Lead creator: Shu-Han Chen, MD, household drugs doctor, chief of the Well being Administration Heart at St. Paul’s Hospital in Taoyuan, Taiwan.
Disclaimer: This text doesn’t essentially replicate the views of AsianScientist or its employees.













.png?trim=0,0,0,0&width=1200&height=800&crop=1200:800)
