Dynaccurate is revolutionising healthcare data in Europe, using AI to clean, structure, and unify medical records, solving interoperability challenges and improving patient safety. CEO and co-founder Dermot Doyle explains.
What specific healthcare problem is Dynaccurate solving, and how is your technology uniquely positioned to address it better than existing solutions?
Dynaccurate focus on medical data refinement, including providing of an Electronic Patient Record that uses AI to clean and structure medical data at source, when written by the doctor. This is to allow health data to become interoperable across the entire spectrum of healthcare, regardless of the system. That’s essentially the problem that all health services struggle with. How do I get the data from one system to another without losing anything and preserving the meaning. If that can be solved, everyone would have 100% of their medical data at the touch of button.
Can you explain in simple terms how Dynaccurate’s technology works and what makes your approach different from other data-driven health startups?
We focus on the data accuracy at source, because that allows for the development of very large data sets which you would use in Clinical Trials or AI training, or underpinning medical AI which gives alerts on patient status. Other health data startups – take your pick! – are typically specific to a certain industry. Some might be making synthetic health data, some might be curating and aggregating data. The sector is much more advanced in the United States than it is here in Europe and beyond.
What measurable impact have you achieved so far, for example, in improving patient outcomes, data accuracy, or decision-making speed?
People still die prematurely from medications which would be flagged if software was connected to one good database holding existing medications data. Our proudest achievement was aggregating all publicly available medications data from across the EU, UK, US and Canada in our www.mdip.eu project. I recently presented results to the European Commission in October this year and proposed that the EU make a single medications database for use by both regulators and clinicians alike.
There’s no reason to split this data and effectively pay for it 10 times in a row, but that’s what we do. We proved that you can take the available data, combine it and enrich with adverse drug reaction data – in this case, pharmacogenomic flags – and make a unitary database for everyone. But what’s happening across the EU is that government, public health and R&D – by which I mean the EU taxpayer who funds this – are paying for the same data over and over again, despite it all being in the public domain. We need a single EU drug index. The EU actually has all the expertise, infrastructure and knowledge. It just requires a push. What we can do at EU level can far outshine anything that national entities can do alone.
How do you ensure data privacy and compliance with European regulations such as GDPR while dealing with sensitive health information?
Currently we don’t hold any live patient data and we generate our own synthetic data instead. We’re using the VANTA to be simultaneously ISO 27001, GDPR and HIPAA compliant for next year. There is also requirements for additional certification inside the EU which will be easy for us to comply with when necessary.
Who are your main partners or clients today (hospitals, governments, pharma companies) and what results have they seen from working with you?
Mostly our partners are collaborators in the research sector, and this is due to the fact that Medtech takes a long time to mature, but also because the complexity of the sector is such that you really need good partners in R&D, particularly if you are going to create a deeptech product using AI. But we’ve also worked with private clients, including clinical, other medtech and the public sector. We’ve done a lot on medications data, like I mentioned before, including building a chemotherapy application for a hospital in Latvia.
But, by far, our www.mdip.eu project is the most impactful, because it cuts through a lot of the tedious ‘interoperability’ discussions that a lot of people obsess over without actually wanting to solve. We said we could build a knowledge graph of medications for the entire EU, and we did it in 9 months. Operationalising that for 27 countries would cost much more of course, but technically there’s actually no barriers at all, and the EU has the infrastructure, data, data model and expertise. It’s just split between the Commission and the European Medicines Agency. That’s the only obstacle.
What is your business model and growth strategy and how do you plan to scale Dynaccurate across Europe or beyond?
Right now, we are looking solely at private sector business models, so private clinics, clinical trials etc. There’s an enormous opportunity here, but it’s really knotty and complex mostly because of legacy technologies and legacy thinking. You can cut through all off that with a next generation platform which is AI native. The biggest issue is behavioural, because new models always encounter pushback. However, the benefits being what they are, change is inevitable. I’m actively seeking a chain of private clinics who are interested in clinical trials to deploy in. I say a chain because what we offer works best with scale, especially for private clinics involved in clinical trials.
Looking ahead, what’s next for Dynaccurate? Are there new technologies, funding rounds, or strategic partnerships on the horizon?
New tech, new partners and new funding is all on the cards. We’re developing a new product for the US market right now which has gotten good feedback so far but is still in ‘stealth mode’. Like a lot of EU deeptech companies, we still actively bid for research funding because it’s necessary for product development and provides important financial de-risking for investors (essentially the R&D funding creates the proof-of-concept and validates it). We’re also involved in defence – in fact our very first contract was with NATO for knowledge graph technology – and there’s a lot we can do in that sector as well. We have a good relationship with VCs, so I know what they want to see, and from our side we’re ready to deploy at scale.
