At a time when the number of cancer cases worldwide is on the rise, radiologists have an increasingly heavy workload. Cancer is a leading cause of death worldwide, accounting for nearly 10 million deaths in 2020, or nearly one in six deaths, reports WHO. Estonian startup Better Medicine comes to the rescue by developing an artificial intelligence-powered system that aims to save up to 13 million man-hours per year.
For 2023 Spring issue, Life in Estonia caught up with the company’s CEO Priit Salumaa.
Could you please describe the beginnings of Better Medicine. How did the idea for the company come about?
At the end of 2019, I left my paid job and started looking for a new challenge during the COVID-19 pandemic. As I was involved in the startup Garage48 hackathon series, Hack The Crisis and Global Hack, I saw that there was a lot of room for digitalisation in medicine and things could be better realised. I was looking for something that would have a positive impact on the world and the medical field seemed to fit the criteria I had set. In the meantime, I had gotten to know Dmytro Fishman, a researcher and now CSO at Better Medicine, Bohdan Petryshak our Chief of AI Engineering, and Martin Reim, a radiologist and now CMO at Better Medicine, and by discussing the issues, we realised that the problems Martin had raised in cancer diagnostics could be solved with the help of artificial intelligence (AI). All in all, the world is currently spending too much time on routine and manual tasks that could be automated.
I have seen someone die of cancer in my extended family, but I also know people who have had cancer and been healed. This is an important subject for me, especially as I have been exposed to it and seen how much suffering cancer causes globally.
We found like-minded people and started gradually. As we delved deeper into the topic, it became clear that there are a number of manual and repetitive tasks for radiologists when measuring tumours. Once they find a tumour, they measure it by drawing lines, and in doing so, the radiologist also has to understand the dynamics of the disease and compare it with the patient’s previous scans. So it dawned on us that all the technical side of measuring tumours could be done by artificial intelligence.
We estimate that automating this work could save up to 13 million hours of radiologists’ time worldwide per year. In other words, 1,427 years of working hours could be saved in a year.
It is also important to note that the WHO predicts the number of cancer cases to increase by more than 27% over the next decade. The burden on both the diagnostic and medical systems will therefore increase, while the number of doctors will decrease. Both problems occur at the intersection of the ageing of society and humanity. Currently, cancer is predominantly a disease of the over 50s, but as society ages, there will be also fewer working-age people. Our work is forward-looking and we want to contribute to these negative trends in order to diagnose cancer more effectively and save society’s resources.
The name of our company reflects very clearly that we want to improve medicine. The output is that we build intelligent tools for doctors to provide better care to their patients. Our potential users are radiologists all over the world who use it in their onco-diagnostic work.
How long have you been in business and what steps have you taken so far?
This is our third year of working as a team, but we set up the company only in late 2020. So far, we have built the first four proof-of-concept models, and started collaborating with Tartu University Hospital and Pärnu Hospital to clinically validate these models. We are building a medical device with artificial intelligence as a key component, and we have now developed, among other things, the software infrastructure to package this intelligence. We have also raised money and set up a digital office in Leeds, England.
These have been successes, but have there been setbacks?
My two previous businesses were in the IT sector and now I’m in the medical sector, and it’s been very difficult for me to get used to how slow processes in medicine are and how traditional the sector is. It is much more difficult to innovate here than in less regulated sectors.
How many companies are there on the market dealing with the same issue, and what are your advantages over them?
There are quite a few companies that deal with a similar subject, but our main difference is that we have a full-body approach. We understand that we need to move as quickly as possible to be able to measure the whole human body because that’s the only way we can help radiologists.
A single organ model: for example, the breast, the lung or the prostate, will not have that effect. Technologically we also have an advantage because we are inventing a new way of training AI, for which we have also filed a patent. It’s very experimental and we don’t know if it will work, but we have a concept for how to develop AI in imaging in the future in a different way than we do now. We have attracted funding from private investors and received a grant from Enterprise Estonia to develop and clinically validate our models. We are currently engaged in research and development, which could lead to a product.
How do you see the current medical tech landscape in Estonia: which companies are involved in necessary and exciting ventures and what are the prospects in this market?
It’s interesting that four years ago there were not very many startups in the medical sector, but COVID-19 gave a strong impetus to their creation. A whole range of new medical companies has emerged in Estonia in the last few years, solving a wide variety of problems, from digital therapies to artificial intelligence solutions like ours.
The whole sector is getting a lot more attention and new things are emerging. At some point, there was a wake-up call leading to the creation of many new companies. Not all of them may succeed because it is a very difficult sector to operate in. But it does give hope that there will be innovation in this sector, which helps to save resources, and that doctors will become more accurate, the sector will move towards more preventive healthcare and people will have more control over their health.
Our hope is that the health insurance system, that we, as citizens, sustain, can continue. Otherwise, there is a risk that the health system will be very grossly underfunded and people will face terrible inequalities in access to treatment and medicine. This is where technology can come to the rescue so that there is less suffering and less inequality in the world.