Worldwide there are 422 million people living with diabetes, according to World Health Organization estimates. The condition can lead complications, including diabetic retinopathy, which can result in blindness. However, a recent study suggests artificial intelligence can help identify diabetic retinopathy early.
Researchers at the Medras Diabetes Research foundation in Chennai India that a smartphone-based device called Remido ‘Fundus on phone’, along with artificial intelligence software, had a high sensitivity for detecting diabetic retinopathy and sight-threatening diabetic retinopathy. In fact, the study, which was published in Nature Eye, found the AI software had a 95 percent sensitivity and 80 percent specificity for detecting any diabetic retinopathy.
“Use of AI to analyze retinal images is appealing as it fits in with the current trend of tele-ophthalmology and telemedicine,” researchers of the study wrote. “Automated diabetic retinopathy grading softwares have potential benefits of efficiency, reproducibility and early detection of diabetic retinopathy happening at the physician level and thus would be useful in reducing the burden to the health systems in screening of the increasing number of people with diabetes.”
Typically patients with diabetes are required to go to regular retinal screenings for early diabetic retinopathy detection and sight-threatening diabetic retinopathy. Typically the exams are done by using a type of fundus camera, but recently smartphone-based retinal imaging has begun to emerge as a cost-effective way to screen the images, according to authors of the study.
“Given the alarming increase in the number of people with diabetes and shortage of trained retinal specialists and graders of retinal photographs, an automated approach involving a computer-based analysis of the fundus images would reduce the burden of the health systems in screening for DR,” authors of the study wrote. “There is hence an increasing interest in the recent past in the development of automated analysis software using computer machine learning/artificial intelligence (AI)/deep neuronal learning for analysis of retinal images in people with diabetes.”
The study looked at a total of 301 patients 18 years old and older undergoing treatment for Type 2 diabetes at a hospital in Chennai, India. Every patient had their retina photographed with the Remidio ‘Fundus on phone’ device. Ophthalmologists independently graded the photographs using the International Clinical Diabetic Retinopathy Classification Scale. The photographs were also graded by the EyeArt AI software. Researchers said the sensitivity and specificity of automated grading were assessed and validated against the ophthalmologists’ grading.
By the conclusion of the trial 296 patients’ retinal images were scanned. Ophthalmologist detected diabetic retinopathy in 191 patients and the AI software detected it in 203 patients. Sight-threatening diabetic retinopathy was detected in 112 patients by ophthalmologists and 146 patients by the AI software.
While researchers said there needs to be more work before recommending AI as a regular use in eye care, they did find the results promising.
“Automated softwares using the AI like the EyeArt along with telemedicine can enable faster real-time screening of large number of people with diabetes,” authors of the study wrote. “Smartphone retinal colour photography combined with an automated detection system can ideally result in models with potential for cost-effective routine clinical use by the primary care physicians.”