X-Ray AI Software Gets Expanded FDA Clearance for Detecting Joint Effusions and Dislocations
Recently published multinational research revealed that AZtrauma had a 95.6 percent AUC for detecting dislocations and a 97.5 percent AUC for detecting joint effusions.
The Food and Drug Administration (FDA) has granted an expanded 510(k) clearance for the X-ray AI software AZtrauma, which can now be employed for the detection of dislocations and joint effusions in adults and pediatric patients > 2 years of age.1
Previously cleared by the FDA for fracture detection,
Overall, the researchers found that AZtrauma had a 98.3 percent AUC with sensitivity and specificity rates of 97.4 percent and 96.4 percent respectively. Specifically, for joint effusions, AZtrauma offered a 97.5 percent AUC, a 96.1 percent sensitivity rate and a 98.8 percent specificity rate. For dislocations, the study authors noted that AI software provided a 95.6 percent AUC along with 93.4 percent sensitivity and 96.7 percent specificity.2
"AZmed's investment allows us to consistently and immediately identify and triage fractures, joint effusions and dislocations across adult and pediatric populations before the radiologist ever opens the study. Our AI is fully optimized to support and prepare the radiologist for the study at hand, aligning perfectly with our mission to integrate image analysis solutions into our all-in-one SaaS platform,” noted Alejandro Bugnone, MD, a musculoskeletal radiologist and chief executive officer of Radin Health.
References
- Radin Health. Radin Health and AZmed elevate radiology workflow with AZtrauma’s expanded FDA-cleared AI capabilities. Available at:
https://www.prnewswire.com/news-releases/radin-health-and-azmed-elevate-radiology-workflow-with-aztraumas-expanded-fda-cleared-ai-capabilities-302779315.html . Published May 21, 2026. Accessed May 21, 2026. - Cohen E, Quertani MS, Beaumel P, et al. Performance of a complete AI radiographic suite across 258,373 X-rays from 26 countries: a worldwide evaluation. Radiography. 2026:32(4): 103361.
















