The US-based research team published a paper in the latest issue of JAMA Ophthalmology that they developed an artificial intelligence (AI) algorithm that can automatically diagnose retinopathy of prematurity. The diagnostic accuracy is higher than that of most professional doctors.
Retinopathy of prematurity is a serious blinding eye disease caused by abnormal proliferation of retinal blood vessels, which is common in preterm infants. The disease is mainly diagnosed by the doctor's view of the baby's eyes and is subjective. In the United States, due to the lack of experienced clinicians, misdiagnosis of this disease can often occur.
To improve this situation, a team of researchers from institutions such as Oregon Health and Science University, Massachusetts General Hospital, University of Illinois at Chicago, and Columbia University Medical Center has developed a new AI algorithm. The algorithm uses deep learning techniques to train through the images of more than 5,000 babies' eyes to learn how to differentiate between healthy blood vessels and diseased blood vessels. Test data shows that after training, the accuracy of the algorithm for diagnosing retinopathy of prematurity was as high as 91%, and a team of 8 retinopathy specialists for preterm infants diagnosed the same image with an accuracy rate of 82%. In other words, the AI ​​algorithm performs better than most professional doctors in the diagnosis of retinopathy of prematurity.
The test population in this study was mainly white infants. The current research team is working with India to test whether this algorithm can accurately diagnose retinopathy of prematurity in India. In addition, the team is still exploring whether the algorithm can detect abnormalities in other parts of the retina other than blood vessels. Researchers said they hope the AI ​​algorithm can be used in clinical practice as soon as possible.
AI applications are currently becoming a hot topic in the medical field. Last month, the U.S. Food and Drug Administration (FDA) approved an AI device for detecting diabetes-related eye diseases, which became a landmark event for applying AI technology in the U.S. medical field. (Reporter Liu Haiying)
Ice Hair Removal,Ice Laser Hair Removal,Ice Therapy Hair Removal,Ice Cold Ipl Hair Removal
Shenzhen Jie Zhong Lian Investment Co., Ltd. , https://www.szmeizon.com