A single retinal photograph contains signatures of cardiovascular disease, kidney dysfunction, diabetes, and early markers of neurodegenerative diseases like Alzheimer's. For decades, we've approached eye disease one condition at a time while research has shown that deep learning can extract 40+ systemic biomarkers from fundus images alone. The signal is there. We just need to learn how to read it and use it.
What we're building
Solayu is an AI research lab focused on ophthalmic imaging. We're building foundation models that work across diseases, across imaging modalities, and across populations. Our models detect multiple diseases simultaneously — trained on diverse global populations, not narrow datasets. We enable comprehensive diagnostics and capture systemic biomarkers through low-cost imaging, helping in early detection, tracking and management of conditions that develop over decades.
Why now
Deep learning works and it's ready to scale. Over the past decade, AI has matched specialist performance in detecting eye disease. More importantly, self-supervised learning has shown that models can find patterns invisible to even clinicians, giving us signals that predict conditions years before symptoms appear.
The need is urgent. 112 million people will have glaucoma by 2040. Half won't know it. In low-income countries, over 90% of cases go undiagnosed. By the time most patients are found, a third already have severe disease. This isn't a problem we can solve by training more ophthalmologists. There aren't enough, and there won't be. We need systems that scale where the workforce cannot.
What we believe
The eye is an undervalued diagnostic site. It's the only place you can non-invasively visualize neural tissue and microvasculature. Every retinal image can yield maximum insight and we're closing the gap between what's shown possible in research and what actually reaches the clinic.