Medios AI is not US FDA approved and not available in the US

eye-g04294be56_1280

The integrated solution provides an instant report in contexts where there are no specialists. (Credit: Tobias Dahlberg from Pixabay)

Glaucoma, one of the world’s leading causes of irreversible blindness, is expected to affect an estimated 120 million by 2040, globally. Treatment options for managing Referable Glaucoma – a stage where immediate treatment improves prognosis, exist, but a simple test to screen has been elusive. Today, the only way to detect is through complex investigations often requiring multiple devices handled by Glaucoma Specialists.

Remidio Inc, alongside Aravind Eye Hospital (AEH), Pondicherry, and Narayana Nethralaya (NN), Bangalore, announced the results of a landmark clinical trial that can help revolutionize the detection of Referable Glaucoma.
The integrated solution provides an instant report in contexts where there are no specialists. It consists of Remidio’s Fundus on Phone, an imaging device that captures images of the retina without pupil dilation, and Remidio’s Medios Referable Glaucoma AI, an integrated offline AI algorithm that requires no internet for inferencing.

The AI looks beyond the cup-to-disc ratio and detects structural changes in the optic nerve head and surrounding retinal nerve fiber layer, critical regions typically assessed by glaucoma specialists during clinical examination with an expensive device called Optical Coherence Tomography.

This video shows how technology can screen for Referable Glaucoma in 15 seconds.

The diagnostic performance of the AI was validated in a first-ever prospective study conducted simultaneously at a tertiary eye care center, NN, and in last-mile access contexts, Vision Centers (VCs) of AEH.

Among the 237 study subjects at NN, the AI results were compared against the current standard of care, wherein Glaucoma specialists provided the final diagnosis following a comprehensive work-up. The algorithm demonstrated a high sensitivity of 93.5% (95% CI: 87.1-96.8%) to detect Referable Glaucoma with a specificity of 85.4% (95% CI: 78.3-90.4%). The recall for ‘no glaucoma’ was also high at 94.7% (95% CI: 87.1-97.9%) with a minimal overcall of normal subjects.

Dr. Rohit Shetty, Vice Chairman at NN, said, “Our clinical study at NN shows that AI can bring much-needed objectivity in glaucoma screening, especially in contexts where no specialists are present. We also see the power of AI to be used in applications as diverse as outreach screening to personalized care.”

In the multi-center study at AEH, among the 299 subjects evaluated at the VCs by residents, 70 were referred to the base hospital. Early results demonstrated a sensitivity of 91.30% (95% CI: 79.2-97.5%) and specificity of 91.67% (95% CI: 73-98.9%). The AI detected Referable Glaucoma with a higher sensitivity than the residents and reduced the number of over-referrals.

The integrated solution holds promise as an easily deployable tool to detect Referable Glaucoma as the results seen in a tertiary eye care center are also seen in last-mile screening contexts.

Dr. R. Venkatesh, Chief Medical Officer at AEH, said, “AI has helped overcome the challenges often presented with Glaucoma screening in the outreach. The reporting of fundus images has gotten much faster and more reliable with AI. It has great promise in VCs as a clinical decision tool for residents to improve their diagnostic consistency.”

Dr. Divya Rao, a Glaucoma specialist and Remidio’s Medical Director, said, “Promising results show that our AI can empower healthcare workers and general ophthalmologists to make an objective screening diagnosis, specifically at centers lacking sophisticated diagnostic equipment.”

Source: Company Press Release