IMPORTANCE OF AI IN OPHTHALMOLOGY.
In the last few years there has been a lot of research and interest about the application of machine learning and AI to analyze medical images to help doctors diagnose or screen patients for diseases due to the fact there is a severe lack of throughput of specialists compared to the number of patients. That is especially true in developing countries.
Diabetic Retinopathy is the third leading cause of vision loss in South Asia. The prevalence of diabetes is estimated to double worldwide, but doctors are limited. In South Asia, the lack of doctors is critical. There are three medical professionals for every 1,000 people in South Asia, while there are 20 professionals per 1,000 people in the West. South Asian countries must solve the issue of access without the economic issues of the West. South Asia has the highest prevalence of blindness and vision loss in the world. Machine learning and AI diagnostics have the potential to fill the specialist gaps for screening our growing population in need of preventative healthcare, especially for diabetes diagnosis, management, and prevention of complications. Globally, the World Health Organization estimates that 217 million people have moderate to severe vision loss, and 36 million people are blind. Diabetic retinopathy as a cause of vision loss has doubled in prevalence since 19901. The rate of diabetic retinopathy is expected to climb, especially in South Asia, as the population ages and the rate of diabetes increases. According to WHO, approximately 80% of vision impairment is avoidable. Diabetic retinopathy is the third most common cause of preventable vision loss. Diabetic retinopathy is irreversible damage to the microstructures of the eye caused by high blood sugar, high blood pressure, and increased blood fats. It is recommended that most diabetics start periodic screening within 2-5 years of diagnosis. For the working poor and those in rural areas, access to eye doctors, the knowledge to access screenings, and the cost of screenings and transportation to screenings are the biggest barriers to preventive eye care.
Remote diagnosis can offer numerous benefits to complement traditional in-person medical care:
- Allows access to specialty care in rural areas without high travel or coordination costs.
- AI diagnostics can offer immediate results from screenings resulting in improved follow-up.
- It expands the geographical reach of a few concentrated specialists.
- Well trained technicians using technology can screen to devote doctors time to critical need patients.
- AI-assisted diagnosis increases a doctors output and allows more time for treatment.
That is where Smart Retina Comes in.
Our Diabetic Retinopathy products, SmartRetina and Retina-AI, connect eye doctors with remote patients while our software keeps permanent visual records. Either system can be easily deployed in remote or underprivileged areas. SmartRetina is compatible with mobile low-cost fundus cameras used by a trained technician. It provides a patient record, data, and images for doctor review. It provides a report to the patient by email or SMS. SmartRetina is an android and web-based platform compatible with many handheld fundus cameras. Retina-AI is a cutting edge AI-assisted screening and diagnostic tool. It immediately generates a probability of diabetic retinopathy, allows magnification, vessel imagining, and color contrast, and highlights areas of interest when reviewed by the doctor. The doctor can quickly and accurately conclude a diagnosis, due to the medical history, advanced imaging, and AI support. Upon the doctor’s approval, an automated report is generated, which can be instantly sent to patients via email and text message. Retina AI is currently undergoing clinical and field testing to be able to complete autonomous diagnostics.