ViewsML Raises $4.9M Seed to Replace Lab-Based Biomarker Staining with AI Pathology Analysis

Pathology has been running the same play for decades. Glass slides, staining cycles, time burned, tissue consumed. It works, sure. So did flip phones. But when speed, cost, and data depth start colliding with modern drug development, “good enough” begins to look expensive.

Now the market is paying attention. ViewsML out of Vancouver just pulled in $4.9M in an oversubscribed seed round led by Wittington Ventures, with Continuum Health Ventures and Mayo Clinic stepping in, and RiSC Capital, Debiopharm Innovation Fund, WUTIF Capital, Defined, and eFund running it back like seasoned operators who know when to double down. Smart money does not chase noise. It follows signal, and this one is getting louder.

Kenneth To is not new to the game. Two decades across medical affairs, product, and commercialization tends to sharpen your instincts or send you packing. Kenneth To stayed and built. Pair that with Christopher Jackson, the mind behind the virtual IHC concept, and George Hennen engineering the backbone, and you start to see the outline of a team that understands both the science and the scale. Not always a given in techbio, where one side usually blinks first.

Here is where it gets interesting. ViewsML is turning biomarker staining into a software problem. No extra tissue. No waiting days or weeks. They are pulling spatial biomarker insights straight from standard H and E slides, cell by cell, like finding hidden tracks on an album everyone thought they already heard. The pitch is simple but sharp. Give researchers back all their IHC time, half their budget, and more data than they had before. Efficiency is nice. Multiplying insight is how markets shift.

The partnerships tell their own story. Dartmouth Health, Providence Health Care, Mayo Clinic. These are not casual handshakes. These are proving grounds. When your product lives at the intersection of AI and pathology, credibility is earned in clinical corridors, not pitch decks.

The takeaway is not just about replacing a lab step. It is about abstraction. When something physical becomes software, margins change, speed changes, and power shifts to whoever owns the model. ViewsML is building a library, not a feature. That distinction tends to separate tools from platforms, and platforms from everything else trying to keep up.

Biopharma, clinical research, diagnostics. If your workflow still depends on processes that have not evolved in decades, this is the part where you start paying attention. Quiet revolutions rarely stay quiet for long.