Q&A: AI software offers cost-effective DR screening without internet access

Key takeaways:

• The Medios DR AI software enables early diabetic retinopathy detection without internet access.
• Studies have validated the AI’s accuracy in detecting DR.

AI continues to change the landscape of retinal disease management regarding diagnosis, monitoring and treatment, and an AI software from India may elevate the field even further.

According to a press release, Remidio, headquartered in Bengaluru, India, and focused on vision health, is harnessing AI to offer widespread cost-effective diabetic retinopathy (DR) screening without the need for internet access. Medios DR AI has been approved by the Central Drugs Standard Control Organisation in India and previously received a CE mark in Europe and approval from the Health Sciences Authority in Singapore.

“It is currently being used in some of India’s most remote areas, from the hilltops of Himachal Pradesh to rural West Bengal, bringing essential eye care to communities that previously had limited access,” the release said.

Divya Rao, MS, chief medical officer of Remidio, spoke with Healio about how Medios DR AI allows for more affordable and accessible DR screening.

Healio: What unmet needs in DR did Remidio set out to address with the Medios DR AI?

Rao: India is home to approximately 77 million people with diabetes, and DR affects nearly one in three of them. Alarmingly, many of the cases remain undiagnosed until severe, often leading to blindness. Compounding this issue is the critical shortage of ophthalmologists — just one for every 150,000 people — making widespread screening by ophthalmologists unfeasible.

Our DR AI was designed to address this gap by enabling early detection through simple-to-use fundus cameras integrated with AI that operates entirely offline. This allows for decentralized screening in remote areas, improving access to care for vulnerable populations and helping to reduce the burden of preventable blindness in India.

Healio: Can you explain how Medios DR AI is able to detect DR without internet access?

Rao: The AI operates entirely on the edge, utilizing the processing power of the smartphone integrated with the fundus camera. With just a click, the AI analyzes the retinal image for signs of referable DR almost instantly, without relying on internet connectivity. Results are available in less than 5 seconds.

This edge-based approach has several key advantages. By processing and storing data locally, it enhances security and privacy, reducing the risk for data breaches. It also lowers cloud-related expenses, making the solution more cost-efficient. Additionally, it maintains operational reliability even in areas with limited or no internet access and supports scalability by distributing the computational load. Importantly, this setup avoids a single point of failure, ensuring that the system remains resilient and effective in any environment.

Healio: How is Remidio approaching the U.S. and other areas?

Rao: We are deliberately taking a multi-AI approach for FDA approval to ensure that our Medios AI offers a meaningful differentiation from currently approved DR AI algorithms in the U.S., aiming for greater patient impact.

Beyond the U.S., we are actively collaborating with groups in the Global South to make our algorithms widely available, focusing on real-world validations to ensure the technology meets the needs of diverse populations. Our goal is to expand access globally while maintaining the highest standards of clinical efficacy.

Healio: Can you explain how the Medios DR AI cuts costs without compromising efficacy?

Rao: The AI utilizes shallow architectures that run directly on a smartphone, eliminating the need for expensive cloud infrastructure while maintaining high performance. Extensive global studies have validated the AI’s accuracy in detecting DR, ensuring that cost savings do not come at the expense of clinical efficacy.

By avoiding the costs associated with cloud storage and processing, the system not only reduces operational expenses but also improves efficiency. This allows for more affordable and accessible DR screening without compromising the quality of the analysis.

Healio: Is there anything else you would like to add?

Rao: We need to rethink how we use AI to improve health care access. It is not just about enhancing existing systems — it is about bringing essential screening services to areas where they simply do not exist.

In developing countries, there is a critical shortage of doctors, and we cannot afford to use specialists for routine screenings. AI can bridge this gap by enabling specialists to focus on complex cases and surgeries, while ensuring that those in underserved areas receive the same level of care through advanced technology. This is the real problem we are trying to solve. Additionally, we focus on the entire patient journey, not just screening. Ensuring care gap closure after screening for those diagnosed with the disease is just as important as AI-led screening.

Reference:

For more information:

Divya Rao, MS, of Remidio Innovative Solutions, can be reached at drdivya@remidio.com.

Published by:

Read more about

This post was originally published on this site