EyeWorld is the official news magazine of the American Society of Cataract & Refractive Surgery.
Issue link: https://digital.eyeworld.org/i/790893
EW NEWS & OPINION March 2017 39 with diabetes: 2002-2009. Diabetes Care. 2014;37:1321–8. Editors' note: Drs. Karth and Rahimy have financial interests with Google. millions of individuals out there who are currently not receiving care right now. Their diseases are not being identified early, and, in fact, by the time they're diagnosed, they may already be in a more advanced state, and the cost related to taking care of patients at that stage con- tributes to the unsustainable health care costs we're dealing with in [the U.S.]. These machines aren't here to replace anybody…there are going to be a lot more people treated." This technology could also dovetail nicely with telemedicine. In the future, this screening algorithm in conjunction with an inexpensive, efficient camera could possibly be set up in public locations, like local drug stores, Dr. Karth said. "With something like this, the machine learning does help minimize the number of steps.… Hopefully, we will get to the point where there will be a cheap, effi- cient, automated camera system. These may be publicly available at a health clinic or pharmacy… taking a picture and at least telling a patient, 'This is referable disease, you need to go now and see a specialist.' Hope- fully it's eliminated several of those steps along the way so the patient can eventually be seen," Dr. Rahimy said. While ophthalmology is a nat- ural fit for deep learning programs, due to its emphasis on imaging, Dr. Karth said, "I think we will start to see machine assistance in more parts of ophthalmology and many parts of medicine in the near future." "There are things—especially repetitive tasks—that machines will just do better than humans," he said. "A machine can look at every pixel of every image, with equal attention, all day and night, never missing a pixel. We are a long way from this kind of algorithm being able to perform as a total screening solution or complete diagnostic aid, but this is one step closer." EW References 1. Gulshan V, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photo- graphs. JAMA. 2016;316:2402–10. 2. World Health Organization. Global report on diabetes. 2016. 3. Bourne RR, et al. Causes of vision loss worldwide, 1990–2010: a systematic analysis. Lancet Global Health. 2013;1:339–49. 4. Shi Q, et al. Racial disparity of eye examina- tions among the U.S. working-age population Creating a new future for your LVC practice. ZEISS ReLEx SMILE // INNOVATION MADE BY ZEISS ZEISS ReLEx SMILE Now available in the U.S. With over 600,000 procedures performed worldwide, the revolutionary ZEISS ReLEx ® SMILE offers your patients an exciting new option and completes your LVC portfolio. www.zeiss.com/us/relex-smile Carl Zeiss Meditec, Inc. 800 342 9821 www.zeiss.com/med REL.8362 ©2016 Carl Zeiss Meditec, Inc. All copyrights reserved. Contact information Karth: peterkarth@gmail.com Rahimy: erahimy@gmail.com