A Fika with: Manuel Marina Breysse, Founder and CEO Of Idoven

Idoven has such a strong mission. Tell us a bit more about it and what prompted you to start out. I’d love to hear about how your background as a successful cardiologist and your contribution to research led you to make the big leap and build a company.

It all started in 2014 with two scientists, a telecommunication engineer and a cardiologist researching machine learning and applying it to cardiology. Jose María (Lillo Castellano), one of my co-founders and I met at CNIC – the Spanish National Center for Cardiovascular Research – one of the top ten in the world. At the time, I was one of the first cardiologists to join the arrhythmia team and Jose María was starting his PhD in machine learning applied to cardiology. We were learning so much, I loved learning about code and its infinite power. 

Back in the day, we would assemble our research into written papers which took us years to publish in high-impact factor peer-review journals, together with our team. Our frustration of not being able to speed up our processes for all the doctors and healthcare professionals dealing with the heavy challenges of diagnosing and helping patients on a daily basis is what really triggered our desire to build a company, together with Iñigo (Juantegui), the third co-founder of Idoven. We decided to automate the diagnosis of five arrhythmias, and after considering how much value this would bring, we were ready to give it our best shot and build a product. Millions of hours are spent in Europe alone, by doctors diagnosing deadly diseases using electrocardiogram (ECG) signals. Transforming repetitious manual tasks into a SaaS platform by training artificial intelligence that supports doctors and patients at scale, will only help save more lives and that has always been our mission. Many challenges we face today in healthcare will be solved through software and AI in the coming years.

Idoven can detect sudden cardiac death many days in advance, it’s truly revolutionary. And as a relatively small team, you’ve worked with some very big and impressive brand names in healthcare. Can you tell us about some of the things you’re already doing?

Today, our AI processes electrocardiograms through our ECG analysis API reducing 98% of overwhelmed cardiologists’ workload. We are supporting pharmaceutical companies in oncology treatments by helping reduce the waiting list of patients starting treatments from one to four months to less than five minutes. 

We recently announced our latest partnerships with AstraZeneca and Google Fitbit, to further integrate into people’s lives. It’s essential these electrocardiograms are recorded through different devices to get the most accurate and extensive data. We’re essentially a layer server that can integrate through an API with any device that is out there, recording heart signals and processing them. I have been building data pipelines since 2012, I know how powerful a data pipeline can be to build a prediction algorithm. And now we are building our own data centre to train all our models. 

This screening process and access to various data points allow us to detect diseases at early stages. Our AI-based ECGs analysis enables us to capture more information from ECGs related to cardiac physiology and pathophysiology than what is currently appreciated by current manual ECG interpretation. We are creating software technology that allows us to measure things that have never been measured before. We’re doing so by working with not only medical device and pharmaceutical companies but also sports organisations such as FIFPRO, tracking the hearts of professional football players to prevent sudden cardiac death. We still have a lot of work and it’s challenging but it’s rewarding to see that our solution can work globally. We’re doing all this thanks to the health data that many patients and research institutions have donated and shared with us.

What you’re doing today is already very impressive and impactful but talking about that next level, I know your long-term ambition is even more impressive. Can you tell us a little bit more about it? 

I think it always comes back to the statistics. Seventeen million people die from cardiovascular diseases every year. A third of those are undiagnosed, meaning they didn’t know they were at risk. Our mission is to reduce the number of undiagnosed patients to zero. We really believe that AI and software solutions will make that a reality over the next decade. It’s what we call cardiology-as-a-service, being able to create a cloud platform that is compliant to operate in many different markets. Beyond this, our prediction algorithm is not just about being able to automate, it’s also about being able to support these professionals with non-human features. 

The prize you’re going after is very large, some early things are really already proving to work. On the other hand, we all know that the medical field, digital health, is an industry difficult to make fast progress in due to all its regulations. How do you navigate that? And in the next couple of years, how do you make sure you get close to that exciting vision fast enough?

The whole company shares the same strategy and culture of value creation. It is so important for us to be able to create value for every single one of the partners we work with, customers, clients, patients, doctors, teammates, providers and so on. Accelerating things like change and changing the status quo is challenging and being in the healthcare system is even more so. If you are a health-tech company, you need to act as a public company years before you actually go public. There are so many regulations and compliance issues we need to meet in order to move forward. We’re working with 15 out of the top 20 research institutions in cardiology in Europe and some of the best hospitals in the US – which is how we got there. Supporting healthcare systems has definitely enhanced our network effect and has allowed us to help those patients that have benefitted.