Imaging artificial intelligence (AI) software developer Imagen Technologies has received U.S. Food and Drug Administration (FDA) authorization to market its AI-based OsteoDetect software for detecting and diagnosing wrist fractures in adults.
OsteoDetect uses machine-learning techniques to analyze posterior-anterior and medial-lateral x-ray images of adult wrists for signs of distal radius fracture. It then marks the fracture location on the image to assist the provider in detection and diagnosis, according to the FDA. The software is suitable for use by clinicians in settings such as primary care, emergency medicine, urgent care, and specialty care such as orthopedics. The FDA noted that OsteoDetect is an adjunct tool and is not meant to replace a clinician's view of the radiograph or his or her clinical judgment.
The software was reviewed by the FDA through its de novo premarket review pathway for new types of low- to moderate-risk devices. The firm submitted a retrospective study to the FDA that compared the independent performance of the image analysis algorithm with the performance of three board-certified orthopedic hand surgeons for detecting as well as localizing wrist fractures on 1,000 radiographic images. Imagen also submitted a retrospective study involving 24 providers who reviewed 200 patient cases.
Both studies found that OsteoDetect improved reader performance -- compared with unaided results during standard clinical practice -- for detecting wrist fractures, providing higher sensitivity, specificity, positive predictive value, and negative predictive value, the FDA said.
