AI Detects Blindness in Preemies with Perfect Accuracy

In a groundbreaking study led by Oregon Health & Science University (OHSU) and its collaborators, it has been revealed that an artificial intelligence technology (AI) holds the promise of flawlessly detecting severe cases of a condition leading to blindness in premature babies. This condition, known as retinopathy of prematurity (ROP), affects the eyes of prematurely born infants by causing abnormal growth of blood vessels near the retina, the part of the eye sensitive to light. The condition is notably responsible for the blindness of famed musician Stevie Wonder.

Published in the prestigious JAMA Ophthalmology journal, the research underscores the AI technology’s unparalleled accuracy in independently identifying all severe ROP instances without the need for ophthalmologist intervention. Annually, ROP puts approximately 2 million babies at risk globally, with the majority of cases resolving without the need for treatment. However, severe manifestations of the condition result in blindness for about 500 babies in the United States and an estimated 50,000 worldwide each year. The scarcity of specialists in low- and middle-income countries exacerbates the issue, leaving many cases undiagnosed and untreated.

“ROP stands as the principal cause of childhood blindness globally, representing a highly addressable challenge in the global initiative to curb preventable blindness,” stated Dr. J. Peter Campbell, M.D., M.P.H., a leading figure in the study, an associate professor of ophthalmology at OHSU’s School of Medicine, and a member of the OHSU Casey Eye Institute. He emphasized the critical shortage of specialists for screening at-risk infants and posited that AI could bridge this gap by enabling efficient bedside screening and directing urgent cases to specialists for necessary treatment.

Originating from OHSU, the i-ROP Deep Learning system employs an AI algorithm to detect anomalies in the retinal blood vessels through image analysis, a task traditionally performed manually by ophthalmologists. This innovation not only offers accurate diagnostics but also facilitates remote consultations, thereby extending the reach of vital eye care services.

This latest study marks a significant milestone as it demonstrates the AI system’s effectiveness in real-world settings, analyzing nearly 12,000 retinal images from over 4,000 infants across neonatal intensive care units in the United States and India. The AI’s performance was exemplary, correctly identifying all severe ROP cases and accurately detecting 80% of cases classified as more-than-mild.

The research and development team behind the i-ROP Deep Learning system includes contributions from OHSU, Massachusetts General Hospital, Northeastern University, University of Illinois Chicago, and the Imaging & Informatics in ROP consortium. The Food and Drug Administration has recognized the system’s potential, awarding it breakthrough status in 2020 to expedite its development.

Looking ahead, Dr. Campbell, who also serves as the CEO of Siloam Vision, the technology’s licensee, is leading clinical trials to further assess the AI system’s efficacy and safety. In partnership with Orbis International, a nonprofit organization, the technology is poised to make significant inroads in improving eye care in low- and middle-income countries.

Should the AI technology gain regulatory approval, it will represent a monumental step forward in autonomous detection of eye diseases, joining diabetic retinopathy as the second eye condition detectable by AI. This advancement holds the promise of transforming the landscape of eye care, offering new hope for early detection and treatment of ROP, and potentially preventing blindness in thousands of infants worldwide.

Source: Medicalxpress

---

Like this article?  Keep up to date with AI news, apps, tools and get tips and tricks on how to improve with AI.  Sign up to our Free AI Newsletter

Also, come check out our free AI training portal and community of business owners, entrepreneurs, executives and creators. Level up your business with AI ! New courses added weekly.