2025-09-10

Artificial intelligence (AI) research and the deployment of AI-based solutions in healthcare are advancing rapidly. AI is positioned as the future of healthcare and can revolutionise patient care by introducing the convenience of care, efficiency, and accuracy in diagnostics for both clinicians and patients alike. However, its benefits have largely been confined to the Global North or high-resource settings in the Global South.

India and other emerging economies are experiencing key epidemiological shifts – an ageing population and a growing burden of non-communicable diseases, particularly diabetes and its associated complications such as diabetic retinopathy (DR). These diseases place a dual burden on health systems already strained by high patient loads from infectious diseases, maternal and child health, and specialist shortages.

AI solutions can help alleviate this burden through clinical decision support and task-shifting. In emerging economies, especially tier-2, tier-3, and rural markets, cloud-dependent AI faces challenges: unreliable bandwidth, prohibitive data costs, and weak digital infrastructure. Offline AI can help bridge this digital divide. With algorithms embedded directly into medical devices and smartphones, it offers a viable alternative for parts of the world with limited or no access to specialists and the internet.

Offline AI for Chronic Eye Diseases

Remidio’s flagship AI, called Medios HI, for diabetic retinopathy, glaucoma, and age-related macular degeneration screening, integrated into its portable Fundus on Phone device, is one of the world’s first ophthalmology AI platforms designed to run entirely on-device, completely offline. The solution is approved by India’s Central Drugs Standard Control Organization (CDSCO), the European Union’s Medical Device Regulation, and Singapore’s Health Sciences Authority.

Crucially, the AI has undergone rigorous clinical validation and has been prospectively validated. A landmark study published in JAMA Ophthalmology in 2019 demonstrated a sensitivity of 100 percent and specificity of 88.4 percent for detecting referable DR in community settings. A prospective validation (Ophthalmology Glaucoma 2024) of the glaucoma AI at Aravind Eye Hospital’s rural teleophthalmology center showed the AI outperformed teleophthalmologists (agreement between AI and glaucoma specialist = 80.3 percent vs 55.3percent agreement between tele-ophthalmologist and glaucoma specialist). Interestingly, another independent validation by Narayana Nethralaya published in Eye (2023) indicated that the AI outperformed glaucoma specialists manually grading images. Glaucoma specialists detected only 60 percent of true glaucoma cases from the retinal images versus 94 percent detected by the AI on the same images, indicating that the AI can detect subtle structural changes not evident to the human eye. With AI, ophthalmologists can minimise the time required to identify new patients and instead devote more time to treatment and direct patient interactions.

Beyond controlled studies, offline AI has demonstrated scalability and cost-effectiveness in real-world deployments across India’s largest eyecare networks.

Implications for Pharma and Med-Tech

For the pharmaceutical and med-tech industries, offline AI represents more than a clinical tool—it serves as a market enabler to identify disease early. Regulatory approval and rigorous peer-reviewed validation provide the assurance required for providers and governments to adopt these solutions at scale.

For pharma companies, these platforms can be integrated into addressing the discovery problem. The economic case for offline AI is equally compelling. Early detection and screening reduce unnecessary referrals and help patients enter treatment pathways earlier. Delivering accurate diagnoses at the primary and community level ensures better retention, improved disease management, and stronger health system efficiency—factors directly aligned with ESG and sustainability priorities of industry leaders.

A Paradigm Shift for Emerging Markets

Offline AI is not merely a technological advancement; it is a paradigm shift in diagnostics delivery for emerging markets. By eliminating reliance on connectivity, solutions such as Remidio’s CDSCO-approved and clinically validated Medios HI have shown that innovation can be both scalable and sustainable. With hundreds of thousands of patients screened, sensitivity and specificity on par with specialists, and demonstrated cost savings in large-scale programs, offline AI is poised to redefine accessibility in diagnostics.

For pharma and healthcare industry stakeholders, this is not only a story of innovation but also a blueprint for scalable, impactful partnerships. Offline AI is set to become a cornerstone in achieving universal health coverage—resilient, inclusive, and transformative in equal measure.