IRIS solves the data bottleneck.

The biology is known.

Thousands of biomarkers are characterized, correlated and published: which proteins rise in sepsis, which combination separates one inflammatory phenotype from another. Decades of work, sitting in the literature.

The models are ready.

Machine learning can read patterns across dozens of markers at once and turn them into a clinical decision. That capability arrived years ago. It is not the constraint.

Nobody has the data.

Generating biomarker data at the volume, speed, and cost the AI models need is too slow and expensive. It requires sending samples out to labs.

IRIS solves the data gap.

One chip, one sample, up to 100 markers at once in minutes.

TX-100 specifications

Representative surface. IL-6 at 6 pg/mm² for 1 ng/mL: Assay Development Gap, 2026. The sub-pg/mm² floor: Mplex, July 2026.

In use at top institutions today.

The IRIS MX-200 benchtop instrument, screen raised and showing binding data.
University of PortsmouthBoston UniversityaxiVENDBerkeley LabUniversity of Texas Medical Branch at GalvestonCNR, Consiglio Nazionale delle Ricerche

What that measurement means for care.

Fast

About ten minutes from sample to result, at the point of care.

Sensitive

At the concentrations that matter, from a single drop of blood.

Multiplex

Up to 100 markers from one sample, read together.

Low cost

A single-use cartridge with the chip inside.

Dynamic range

Over four logs, enabled by kinetic precision.

TX-100 · in development · not for saleNot for use in diagnostic procedures.

Where this technology first changes diagnoses.

The first panel is in development now. Every one after it is the same chip and the same workflow carrying a different set of proteins, which is why the list grows without new hardware. What follows is the plan, not a schedule that has been met.

Stacked area chart of the planned diagnostic panels growing over four phases on the same instrument: Trauma first, then Sepsis, Cardiac, Neuro, and more.
And more
Neuro
Cardiac
Sepsis
Trauma
From phase 4, all on the same installed instrument.
Phase 1
Trial sites and labs
Phase 2
510(k) for biomarkers
Phase 3
Market launch
Phase 4
Expansion

One technology. Three form factors.

The sensor that measures binding kinetics on a lab bench is the sensor that goes into a clinical cartridge. Nothing is reinvented between them, which is why the research business and the diagnostics business are one company. Nothing here waits on a breakthrough either: the chips come off standard semiconductor production lines, the light source runs on off-the-shelf LEDs, and a standard CMOS camera reads the array. Scaling up is a manufacturing job, not new science.

The IRIS MX-200 benchtop instrument.

MX

MX means multiplex.

The research instrument. Develop the assay, characterize the binding, prove the panel. This is what ships today.

Shipping · research use onlyThe MX instruments
The IRIS TX-100 development prototype.

TX

TX means tabletop.

The same chip in a single-use cartridge, read at the point of care in about ten minutes. This is what is being built.

In development · not for saleThe TX-100

HX

HX means handheld.

The same measurement, wherever the patient is. This is where it goes next.

Roadmap

The letters are ours, not the field’s. IRIS MX-series instruments and IRIS Chips are for research use only and are not for use in diagnostic procedures. The IRIS TX-100 is in development, is not available for sale, and is not for use in diagnostic procedures. This page describes IRIS Kinetics’ technology and plans. It is not an offer to sell, or a solicitation of an offer to buy, any security. Statements about products in development and future plans are forward-looking and may change.

A panel you can only run in a lab is a panel you run rarely, and never in the hours that decide a case.

IRIS is building the same measurement into an instrument that goes where the decision is made.

An IRIS chip, photographed: a blue 8 by 16 mm silicon chip with an alignment arrow, a printed scale grid, a QR code and the IRIS mark.