AI Algorithm That Helps Doctors Diagnose Wrist Fractures Approved by FDA

AI Algorithm That Helps Doctors Diagnose Wrist Fractures Approved by FDA

The U.S Food and Drug Administration (FDA) has approved a new AI algorithm to help physicians diagnose wrist fractures in adult patients.

The computer-aided detection and diagnosis software platform – OsteoDetect – was developed by Imagen Technologies and has been trained to detect distal radius fractures, a common type of wrist fracture, from X-ray images.

2D X-ray images are fed into the system which checks the images for signs of fractures. When a fracture is detected, the location is marked and the X-ray images are then assessed by a trained human eye.

It is hoped that the new AI-based software platform will help healthcare providers detect wrist fractures faster and that it will prove to be an invaluable tool to aid the diagnosis of wrist fractures. The FDA says the system is likely to be used in a wide range of settings including primary care facilities, urgent and specialist care facilities, and in emergency medicine.

The FDA approved the AI-based software through the De Novo premarket regulatory pathway for low to moderate risk devices.

While the software can help to diagnose wrist fractures more quickly, it is not intended to replace clinicians. A review of radiographs by a clinician will still be required.

Imagen Technologies conducted a retrospective study of 1,000 radiograph images and assessed the performance of the algorithm for fracture localization against three board-certified orthopedic hand surgeons. A retrospective study of 24 providers who had reviewed 200 patient cases was also submitted.

In both studies, the FDA reports that the detection of wrist fractures was improved using the OsteoDetect platform, increasing sensitivity, specificity, positive and negative predictive values.

Now that FDA approval has been received, Imagen Technologies has the green light to start selling the technology to healthcare organizations.

Leave a Reply