Everyone guesses. Everyone loses.
lost every year to cassava disease in East & Central Africa alone.
of global crop yield destroyed by pests & plant disease, every year.
annual cost to the global economy (FAO estimate).
This isn't an edge case. It's a structural, recurring tax — paid in food and livelihoods by the people who can least afford it.
Every one shares the same root cause: a decision needed in situ — but data only available from a distant lab, days later.
Same four domains as the problem — now with the risk turned down, in situ, in minutes.
Not one disease. Any biomarker. That's the company we're building — cassava is simply where we prove it.
Imperial PhD (Güder Lab). ELLA co-inventor; owns the science, v1 product & manufacturing, next-gen impedance platform (patent pending).
Biomedical engineer bridging Imperial deep-tech to market. Owns BD, fundraising, IP/data governance, and the entire live software stack.
Leads the Güder Lab, a world-class paper-diagnostics group. Holds the Gates relationship, grant base and lab resourcing.
Behind us: José Flauzino & Laura Gonzalez Macias (lead inventors), an Imperial wet-lab bench, partners IITA · IFAD · GAIN.
ELLA turns any lateral-flow test — the most universal diagnostic format on earth — into a quantitative, GPS-tagged digital read. Battery-free, NFC tap from any phone, sub-$1 per test. Biomarker-agnostic by design.
2 granted Imperial patents · Biosensors & Bioelectronics (2024) · Flauzino et al. 2025 preprint.
The ELLA app captures each read — result in under a minute, geotagged, timestamped — and syncs to the cloud. The dashboard aggregates thousands into a live surveillance picture a programme can act on.
Live today at dashboard.superposition.bio — multi-tenant, the centrepiece of our Gates demo. This stack is built, not promised.
Cheap quantitative reads at scale → chemical datasets no one else can assemble. Every biomarker a new dataset; every tap a new row.
ML fusing diagnostics with weather, soil & agronomy to predict risk and recommend the action — spray, accept/reject, treat/don't.
Replication cost ≈ zero → scales at AI/data multiples (6–10×+). We sell certainty, not cassettes.
The device is the on-ramp; the data layer is the defensible asset — and it compounds with every deployment.
More volume of the same use case — every new cassava site, buyer and region. Same product, more reads.
More use cases. The read is already proven across 7+ biomarkers — banana, sepsis/AMR, folate, respiratory, ferritin/Vit D. One method, four domains.
The data layer compounds — more reads → sharper predictions → new markets for intelligence and a self-improving reduction of biological risk.
The only platform at the convergence of four large, fragmented domains plus a compounding data layer. Headline TAM ~$60B today → ~$95B by 2030. The opportunity is huge.
To commercialise CBSD detection and fund the first high-value expansions.
$8M so farmers like Linda don't lose a crop to a disease they couldn't see. One platform. Countless biomarkers.
We're also opening a handful of pilot partnerships — warm intros to anyone wanting to turn a biological blind spot into a measured, managed risk are welcome.
Thank you. · Thank you.