Eqvista sat down with Antxon Caballero, CEO of Heuristik, to explore how a personal family crisis sparked a groundbreaking biometric solution for patient identification in healthcare. Heuristik’s origins stem from a near-fatal medication error, highlighting Europe’s widespread patient misidentification issues affecting up to 12.5% of cases on average.
Antxon explains the company’s AI-powered fingerprint technology, which enables rapid, fraud-resistant patient verification even for unconscious or non-verbal individuals, outperforming traditional wristbands or cards.
Heuristik integrates biometrics with real-time analytics to track patient locations and workflows, reducing errors, duplicates, and bottlenecks while ensuring GDPR compliance through proportional data use without storing personal information. The startup has successfully piloted in hospitals, published evidence of 100% accuracy in real-world settings, and raised €2.1M in pre-seed funding from investors such as SETT Spain and Lemon Ventures.
Looking ahead, Heuristik aims to expand across Europe, the US, and Mexico in 2026, targeting certifications and positioning itself as the standard for patient safety and management.
Antxon, your company was born from a deeply personal experience. Can you share how that crisis became the catalyst for what you’re doing today, especially given the massive scale of patient misidentification problems in Europe?
Sure, this happened about 6 years ago, at that time I was taking care of my grandparents. Basically, I received a phone call from my grandmother saying that a nurse just gave the drugs to my grandfather and they were going to start providing the eye medicines. I told my grandmother that no one was supposed to provide any drugs to my grandfather, so I asked her to put me in contact with the nurse.
The nurse said that she already provided the drugs to Gilberto, and I asked her who Gilberto was… she answered that Gilberto is the persona that is in front of her and I replayed that, this was my grandfather, called Trinidad, then she made the most professional move I’ve ever seen, she hangs out the phone and we never knew anything else. My grandfather started to feel bad and almost cost his life. At that moment, I started researching patient identification errors and discovered. I just found the top of the iceberg.
Once you identified this problem, what made you believe biometric identification combined with AI was the right solution, rather than improving existing systems like patient wristbands or electronic health records?
At the beginning even before starting to develop the solution I started talking with hospitals and some of them told me that they tried to use a biometric solution but didn’t work, so that’s when I thought about creating an AI model able to solve the problem. We recently published a paper (Preprint stage), where we have identified 100% of the patients correctly in real healthcare settings. Before starting with the pilot, one of the technical leaders of the hospital told us that they tried a fingerprint solution in the past that didn’t work and she wasn’t totally sure about our solution, but after the pilot she was convinced that our breakthrough solution had potential.
On the technology side – your solution combines fingerprint biometrics with AI. Can you walk us through how it works in practice? For instance, when an unconscious patient arrives at an emergency room without ID, what’s the user experience for the healthcare worker, and how quickly can you identify that patient?
Currently, when a patient arrives at a healthcare entity in admission, healthcare staff identify the patient using a healthcare card or identification card, after that they will put a bracelet on the patient. At that stage, there is fraud, problems with providing the correct bracelet to the patient, etc.. and situations where they cannot identify the patient correctly. For example, one of the hospitals that we are working with reported that around 35% of the patients cannot follow the current patient identification protocols, because they cannot answer during the whole patient journey their name/surname and date of birth.
Using Heuristik, healthcare staff are going to be able to identify the patient during the whole process using their fingerprint. No more fraud, doubled medical records, problems with patients with the same name and surname.
How did you secure your first hospital pilot for a biometric patient identification system, given that healthcare providers rarely allow startups to experiment with such a critical safety process?
Because there is an unsolved need for a solution that can improve patient safety. The average patient identification error statistic is around 12,5% and on average hospitals lose around 17% in patient identification error situations. We work for healthcare staff, we want to make their job easier and help them identify patients in a better way, so they can focus on the most important thing, patients. And the most important part is that we are creating scientific evidence with papers to demonstrate that using Heuristik improves patient safety and is cost-efficient.
How do you balance the need for biometric data with stringent privacy requirements like GDPR? Where is the data stored, who controls it, and how do you ensure patient consent?
We are GDPR compliant because we never have access to patient information and the identifications are proportional this means that there is not another way to make those identifications or healthcare entities are struggling in those areas.
Beyond fingerprint-based identification, how do you see Heuristik evolving as a data and workflow platform for hospitals over the next five years?
Heuristik is not just about identifying patients we also provide real time statistics to healthcare entities in order to avoid bottlenecks (Where the patient is located in real time, how much time it takes to make any proces). We are working to become the standard of patient identification and management.
What’s your strategy for growing internationally? How do you handle the different healthcare systems, regulations, and purchasing processes across countries?
In 2026 we are going to be implemented in a couple of hospitals around Europe, the US and Mexico. Even in the same country healthcare entities are different so there is not an issue for us, as soon as they find an area that we can help improve, we are ready to go.
What regulatory or certification milestones in Europe and the US are true ‘unlock’ moments for your go-to-market, and how are you sequencing them?
In Europe, GDPR is a big thing, so you need to be working on this for some months, specially dealing with sensitive data. During 2026, we are planning to achieve some major certifications in Europe, but just to build trust around information security.
Congrats on raising €2.1M in your pre-seed round. What was the hardest part of that process, and how did it feel to have investors like SETT Spain, Fundalogy, Startupxplore, and Lemon Ventures support your vision?
The hardest part in general is to be a Digital health/ Deeptech company. Because healthcare investors are not interested in companies that are not medical devices or life science and deeptech investors/conventional investors do not want to invest in solutions that sell to hospitals.
So, in order to build credibility we had to close hospitals, build a strong team and show them that we were able to achieve the KPIs we promised.
Finally, if you could give one piece of advice to a solo founder working on patient safety tech, would it be about the technology, the market, or something else?
Focus on your goal and it doesn’t matter what happens following your goal.
