Breaking the Vicious Cycle of
Blindness in Djibouti

Djibouti8 LG -WOMAN

A patient in the Djibouti General Hospital awaits the results of retinal photography. Fortunately, she didn’t have retinal disease.

By Eric Craypo*

Here’s the catch: a person with Diabetic Retinopathy – the leading cause of blindness in working age adults around the world – can still see well when they are in the early stages of the disease. But if they wait until visual symptoms appear to seek help, it’s often too late;and the treatment is unlikely to work. Then as vision becomes further impaired, they blame the treatment for the blindness, which discourages others from seeking treatment. It’s a vicious cycle of blindness.

The solution, explains Berkeley Optometry’s Dr. JorgeCuadros is early detection and patient engagement: “If you treat it early, 90% of people can maintain good vision. You can avoid vision impairment.” But in many places around the world – including underserved communities in this country – the screening of diabetic patients is hampered by both a dearth of equipment and clinicians trained to do the assessments. Djibouti, for example, has only two ophthalmologists for the entire country.

With this in mind, Dr. Cuadros traveled to Djibouti, an east African country of 800,000 located on the Red Sea,late last year to train a group of nurses and other clinicians to operate a digital retinal camera and to detect Diabetic Retinopathy using a free and non-proprietary software called EyePACS—a program developed by Dr. Cuadros and Dr. Wyatt Tellis of UCSF.

Dr. Ethan Chorin, founder and director of Perim Associates,the Berkeley-based international policy consultancy that helped introduce EyePACS in Djibouti, says: “EyePACS is a great example of technology in service to development. It is simple to use, responds to a widespread need, and empowers local clinicians to do their own screening and treatments, rather than relying on external aid.”

retina01

Retina of diabetic
diabetic retinophaty

During the initial screening of 140 people, the group found that 64 had some diabetic retinal disease and will need to be counseled and monitored closely to avoid vision impairment in the future, and 20 had severe diabetic retinal disease requiring immediate treatment. For these84 people, the screenings have likely prevented blindness.But it was too late for 5 of those 20 patients—the disease had progressed so far that treatment is unlikely to help. A devastating reminder that early detection is critical. In the next phase of the project, Cuadros anticipates that newly trained health care professionals – traveling around the country—will be able to screen 50% of Djibouti’s diabetic population. That’s about 40,000 people. He thinks it can be done by the end of 2017. The program will follow the success of similar endeavors that Cuadros initiated in the U.S., Canada and Mexico. All told, over340,000 people with diabetes have been screened using the EyePACS system, saving thousands from blindness. Dr. Cuadros recently told the Huffington Post, “If we can test early and widely, we can save many from this fate. The testing technology is now there, it’s simple, portable, and it works really well.

*Originally Published in U.C. Berkeley Optometry Magazine, November, 2015

Artificial Intelligence
Boosts Diagnostic Power

By Dr. Jorge Cuadros
Last year the California Health Care Foundation (CHCF) sponsored a $100,000 competition using the Kaggle data science community to find the best algorithms, or computer programs, for detecting diabetic retinal disease from digital images. In all, 661 teams of engineers and researchers throughout the world submitted 6,999 algorithms, calibrated using 100,000 images captured through the EyePACS system (see main article). By the fourth month of the competition, algorithms were matching the diagnostic performance of humans. Six months later, Professor Ben Graham of the University of Warwick (UK) won the competition with an algorithm that performed “better” than typical human graders, with an accuracy score of 86%. The remarkable results of the Kaggle competition show that computers can indeed detect sight-threatening diabetic retinopathy better than
human experts, but it is just a step towards developing technology and systems that can truly get people into treatment early enough for the it to be most effective. Many, if not most, patients will seek treatment for diabetic eye disease only when the disease has progressed to the point where effectiveness of treatment is reduced by 50%. Many patients wait too long to be treated, even when they know they have the sight-threatening condition. We should not expect that an app that shows a green light for “good retinas” and a red light for “bad retinas” will make much difference in preventing blindness from diabetes: patients still must be advised, educated, and cajoled to take charge of their health and overcome fear, mistrust, misinformation, cost, and other obstacles to seeking care. Unfortunately, there isn’t an app for that — at least not yet!

The full piece is available in the AR3 Q2 issue.

Leave a Comment

Contact Us

If you have any questions regarding AR3 Magazine, send us an email and we'll get back to you, asap.

Not readable? Change text. captcha txt
Yale_Climate_Backgroundearths-legacy-cover