AI Software Detects Osteoporosis from Routine CT Scans

By that point, bone loss may already be advanced. Bone density testing is not typically performed until osteoporosis is suspected. To help detect the condition earlier, researchers are developing software that measures bone density from existing computed tomography (CT) scans.

The software, called BMDNow, is being developed by the Fraunhofer Institute for Computer Graphics Research IGD (Darmstadt, Germany) with Goethe University Frankfurt and Garritz Online Media International. It measures bone mineral density in individual vertebrae using CT images acquired for other purposes. This approach does not require an additional scan or specialized equipment.

BMDNow uses neural networks to identify the trabecular region, the spongy tissue inside each vertebral body. An imaging algorithm then analyzes bone components, including fat and calcium, and produces a three-dimensional, color-coded map of density distribution. Calibration curves generated with reference phantom models are used to calculate bone density values that indicate osteoporosis or osteopenia, a precursor to osteoporosis.

The team tested artificial intelligence models under real-world conditions using patient data from Frankfurt University Hospital. The researchers are currently using dual-energy CT scans, which acquire images at two X-ray energy levels, to measure density in individual vertebral bodies. They plan to adapt the algorithms for conventional single-energy CT scans, while finalizing and validating the prototype. Their longer-term plans include a reference database, with medical device approval targeted within one to two years.

“In contrast, our software uses existing routine CT image data—such as from heart or lung scans—without the need for additional specialized examinations. This reduces patients' individual radiation exposure and saves physicians' time,” said Stefan Wesarg, a research scientist at Fraunhofer IGD in Darmstadt. 

“Our algorithm is applied specifically in this inner region of the vertebrae to generate a three-dimensional map of bone density distribution. The AI model analyzes bone components such as fat and calcium and uses color coding to visualize their proportions,” added Wesarg.

Related Links
Fraunhofer IGD