Most hospital data is written down, not structured

Healthcare has been digitizing for decades, moving from handwritten charts and carbon copies to electronic health records. But digitizing a record is not the same as structuring it. Most of what a clinician writes ends up as free text, not the ordered, queryable data that clinical research or resource planning actually needs. That gap forces hospitals into a bad choice: shelve research projects that depend on that data, or pay people to structure it by hand, a slow, expensive task that pulls doctors away from both patients and research. Unstructured data does not just sit there unused. It actively delays medical advances that depend on being able to analyze what already happened to real patients.

Natural language processing built specifically for clinical text

IOMED's answer is a natural language processing engine built specifically to structure free-text medical records at scale, something that is harder than it sounds given how syntactically dense and specialized clinical language is. The system standardizes what it extracts into the Observational Medical Outcomes Partnership (OMOP) Common Data Model, an open standard used across the healthcare research community, and it runs on-premise, keeping data inside a hospital's own systems and under the hospital's own control rather than moving it externally. That combination, real structuring of previously unusable text plus a privacy-preserving, standards-based output, is what lets hospitals put historical records to work for research without compromising on data governance.

From one clinical research problem to a healthcare data company

Founded in Barcelona in 2016 by Gabriel Maeztu, a medical doctor who ran into the data-structuring problem firsthand while trying to manually collect data for a clinical research project in a hospital, IOMED is built directly around solving that same problem. He built the company together with Javier de Oca and Álvaro Abella, and IOMED launched shortly after, first deploying in hospitals around Barcelona before expanding across Spain. IOMED is part of the Observational Health Data Sciences and Informatics (OHDSI) network, the open-source community built around the OMOP standard, reflecting its role as an infrastructure layer for evidence-based medicine rather than a single-hospital tool. The company is expanding its footprint among healthcare partners in new markets.

A founding doctor as CTO, and a chief executive out of life sciences

Rohit Mistry serves as Chief Executive Officer, having joined with more than fifteen years in the life sciences industry. Gabriel Maeztu, who founded the company, is its Chairperson and Chief Technology Officer.

When we invested

We first backed IOMED in 2020, participating in a €2 million round alongside existing investors Easo Ventures and SpeedInvest. It was our second investment in digital health, following Quibim, and it reflected a broader thesis that enterprise-grade technology had real room to run in modern medicine, an industry at the center of a genuine data problem. We returned for IOMED's €10 million Series A, led by Philips Ventures, the venture arm of health technology group Philips, and XTX Ventures, the venture arm of algorithmic trading firm XTX Markets, alongside Fondo Bolsa Social, Redseed and EASO Ventures. The company separately received a research grant of close to €1 million from red.es to support a new R&D project in AI.

Where things stand

The company has used the years in between to move from a Barcelona hospital pilot to a broader healthcare data infrastructure play built around the OMOP standard. IOMED is pushing into new markets, and we back the team's aim of building a more collaborative, evidence-based healthcare ecosystem grounded in real-world data rather than manually curated samples.