AI software could detect osteoporosis in existing CT scans before bones break
Osteoporosis progresses gradually and painlessly over many years. Many people do not have their bone density measured until they break a bone or develop back pain. By then, bone loss is usually already well advanced. In the BMDNow joint research project, researchers at the Fraunhofer Institute for Computer Graphics Research IGD are collaborating closely with Goethe University Frankfurt and IT…
Osteoporosis, a condition that gradually weakens bones over many years, often goes unnoticed until a fracture occurs or back pain develops. Current diagnostic methods typically involve specialized tests, like dual-energy X-ray absorptiometry (DXA), which are performed only when bone loss is suspected. However, Fraunhofer researchers are developing AI software that could detect osteoporosis in existing CT scans, reducing radiation exposure and improving healthcare efficiency.
The BMDNow software, created through a joint project by Fraunhofer IGD, Goethe University Frankfurt, and Garritz Online Media International, uses existing CT scans to precisely determine bone mineral density (BMD) of individual vertebrae, without the need for additional staff or equipment. The algorithm employs neural networks and advanced imaging methods to segment the spongy trabecular region of vertebrae, generating a 3D map of bone density distribution.
By analyzing bone components like fat and calcium, the AI model calculates bone density using calibrated values from reference phantoms. This early detection enables timely treatment. The software currently uses dual-energy CT scans but aims to apply to single-energy CT scans in the future, with the goal of establishing a reference database.
The prototype is currently being validated, and medical device approval is expected within one to two years. The project, which won second place in the Hessen Ideen competition in 2025, aims to reduce individual patient radiation exposure and improve healthcare system efficiency.
Written by urgent.news from Medical Xpress's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
