FDA Clears Artificial Intelligence Software for Wrist Fractures

Less than two months ago, FDA commissioner Scott Gottlieb, M.D. announced that the agency would be “implementing a new approach” to the review of artificial intelligence (AI)-based medical devices and tools and digital health as a whole.

Weeks after the announcement, Imagen Technologies, a three-year-old company, received FDA de novo clearance to market its OsteoDetect computer-aided software, designed to detect and diagnose adult wrist fractures. (De novo is a regulatory pathway for low to moderate risk devices of a new type. Moving forward, AI technology will be included in FDA’s Pre-Cert Program.)

This news caught our eye for a couple of reasons. First, we found the OsteoDetect software to be innovative. It uses an artificial intelligence algorithm to analyze 2D x-ray images for signs of distal radius fracture. The software marks the location of the fracture on the image to aid the provider in detection and diagnosis. It analyzes wrist radiographs using machine learning techniques to identify and highlight regions of distal radius fracture during the review of posterior-anterior and medial-lateral x-ray images of adult wrists.

OsteoDetect can be used in various settings, including primary care, ER, urgent care and specialty care. It can also support early intervention of disease.

As we know, innovation must be backed by data about clinical effectiveness. On that note, Imagen submitted results from a retrospective study of 1,000 radiographic images that assessed independent performance of the image analysis algorithm for detecting wrist fractures, and the accuracy of the fracture localization of OsteoDetect against the performance of three board-certified hand surgeons. The company also submitted a retrospective study of 200 patient cases reviewed by providers. Both studies demonstrated distal fracture detection with use of the software vs. unaided performance.

Second, we found the backgrounds of Imagen’s executive and clinical teams to be of interest. The company was co-founded in 2015 by doctors from the Hospital for Special Surgery, scientists and engineers.

Imagen has a robust list of clinical advisors with diverse orthopaedic backgrounds, including knee, hip, shoulder, foot/ankle and trauma. This leads us to believe that the company could expand its technology and software to other applications beyond wrist fracture detection. Imagen’s clinical team comprises radiologists and surgeons from Hospital for Special Surgery, Mayo Clinic, Stanford Hospital, Memorial Hermann, Mount Sinai, Harborview Medical Center and others.

Additionally, the company has hired personnel with experience building technology and products at companies that include Google, Facebook, Etsy and Dropbox. Imagen’s Head of Regulatory Affairs and Quality Assurance, Ninad Gujar, Ph.D., previously worked at Samsung Healthcare as Head of Regulatory and Quality.

The company has raised $21 million in funding to date. According to its website, Imagen’s long-term vision is to “keep patients healthy by delivering novel insights to support early disease identification and improved patient management across episodes of care.”

Given Imagen’s personnel and unique approach to diagnostic care, we are interested to see what other technologies the company brings to the market. Further, we expect other companies to continue investing in AI healthcare solutions, especially if the regulatory pathway is becoming easier to navigate. Consider Gottlieb’s recent comments on where FDA stands on AI: “[It] holds enormous promise for the future of medicine…we expect to see an increasing number of AI-based submissions in the coming years…these are no longer far-fetched ideas.”

Editor’s note: Imagen is one of several companies applying algorithm-based technologies in varying ways for use in orthopaedics. Others include DePuy Synthes, EnHatch, Numerate, NuVasive and Zebra Medical Vision.


Rob Meyer is ORTHOWORLD’s Senior Editor. ORTHOWORLD’s Editorial Assistant, Julie A. Vetalice, contributed to this report.

Join us!

The best of BONEZONE content delivered to your inbox, twice each month.

RELATED ARTICLES



CONTACT BONEZONE

 

CONTACT BONEZONE