AI is making remarkable progress in the healthcare sector, with its potential applications in various aspects of the healthcare sector, such as clinical decision-making, healthbiomedical research, drug development, health system administration and service redesign.
While many countries around the world seek to become early adopters of the use of AI in this sector, Croatia is among the most advanced countries. pioneers in the CEE region to leverage AI to advance its healthcare landscape.
In recent years, efforts in the country have been led by AI4SANTE, a nonprofit consortium of experts in AI, healthcare, and startups. It all started when researchers from the consortium saw an opportunity for Croatia to create what is called a European Digital Innovation Hub (EDIH) for the application of AI technology.
“We realized this was a great opportunity to create a hub that revolves around our experience in AI and how the technology can integrate into healthcare and medicine, which which I think is probably extremely timely and very useful for moving medicine forward more quickly,” Anja Baresic, coordinator at AI4Health, tells The Recursive.
Across the EU, these digital innovation hubs are designed as a supporting entrepreneurial infrastructure that helps businesses, the public sector and other stakeholders in the process of business digitalization, digital skills development and mediation between service providers in the field of new technologies and users of the sector. field of artificial intelligence, cybersecurity and high performance computing (HPC).
Baresic herself has extensive experience in machine learning and computational biology and is a researcher at the country’s largest scientific research center, the Rudjer Boskovic Institute. Together with other researchers and members of AI4HEALTH, Baresic initially saw as a challenge for Croatian healthcare that, although the development of technology is quite rapid, there is a deficit on the application side.
“There are two outliers that make things really difficult. On the one hand, most industries and companies shy away from the healthcare sector if they develop AI. On the other hand, some are hesitant to use AI if they work in the healthcare sector. So we thought that the vast majority of these companies were probably experiencing the same problems and we needed to help them solve these challenges much more quickly,” says Baresic.
What AI4HEALTH then did was initiate the creation of a sort of digital health ecosystem.
“We have brought together all the players in the digital healthcare sector in Croatia. We have both the regulatory side, the tech companies providing the infrastructure, the academia, the faculties teaching, the incubation and access to funding. And we also have a few healthcare companies that want to serve as a testing ground for these services,” Baresic points out.
Practical use of AI in Croatian hospitals
Some of the AI solutions developed by the startups are already adopted in Croatian healthcare institutions.
One of the use cases is a text-to-speech solution developed specifically for the Croatian language and trained on medical records, developed by a Croatian AI startup, Newton Technologies Adria.
As Baresic explains, the solution is used in more than 200 general and hospital practices across the country.
“What they did was develop tailor-made speech-to-text synthesis for medical texts in the local language, which is obviously quite difficult because those languages are quite complex. And what that offers the clinician is a service where they speak into a dictation device, and that is translated into written text. And after a brief check and correction, it speeds up the process of entering a medical report by a factor of 8 to 10. It is very useful and so far it has been used a lot,” Baresic tells The Recursive.
Another practical case involves using a wearable ECG (electrocardiogram) device and AI solution to improve the diagnosis and treatment of atrial fibrillation or irregular heart rhythm. The device, which combines Slovenian technology and the AI solution developed by Polish startup Caridomatics, has the ability to record up to seven days of signal, while traditional recording lasts only 24 hours.
“The idea here is that if you can catch it early and treat it, the preventive value is enormous. The facility also has high potential for use in primary care, as the GP does not need to have any knowledge of cardiology and does not need to refer to a cardiologist based on a very small amount of data. So you provide them with a nice little report where they are much more informed – moving the more complex tasks from secondary practitioners to the GP means it becomes more accessible. For example, in rural areas, it’s always difficult to know if you’re going to send someone to the nearest cardiologist who is miles away,” Baresic tells The Recursive.
Then there is also the use of conversational AI. For example the Croatian startup Mindsmiths and its chatbot Megi was launched in late 2020 and integrated into local hospitals where it helps chronically ill patients with hypertension.
“The chatbot helps them track their blood pressure and heart rate, it feeds them into the system and collects data, then exposes it to the cardiologist on the other side. It also guides patients by allowing them to actively participate in their own health care, which is probably the only way forward for chronic diseases,” emphasizes Baresic.
The potential for increased use of new AI solutions
Umo neuroscience is a Croatian startup that develops neurofeedback and biofeedback technologies and processes. According to its CEO and founder Marin Maletic, the use of AI in healthcare will reach new heights in the coming period.
“We see this clearly in our business of developing and delivering innovative neurofeedback services, which sit at the frontier of healthcare and wellness. We are, as we speak, working on the use of AI in sales, customer management, neurofeedback protocol development, EEG signal processing – virtually everywhere we look, we let’s consider AI as a potentially great tool for improvement,” says Maletic. The recursive.
However, there is also room for improvement, and according to him, this could be achieved by adopting a more humanistic approach to AI, as some Croatian companies are already doing.
“Many people feel alone and confused when they receive a serious diagnosis. This actually happens with all kinds of health problems, but when it comes to life-threatening illnesses, it is a cause of great pain and fear. Health systems inevitably treat humans as statistics, without having the resources to make them agents of the healing process. Mindsmiths has done great things in this area, I think they are a great example of a humanistic approach to AI which I generally favor a lot,” adds Maletic.
In this sense, according to Baresic, AI technology was never intended to replace doctors, but only to help them do their jobs better and spend more quality time with their patients. According to Deloitte reportAI applications have the potential to free up between 1,659 million and 1,944 million hours each year.
“Studies show that you should always have a 20-minute exam with your specialist. However, these days you’ll probably only have two and a half minutes where the doctor has to ask you questions based on all their other thoughts – they have to do both physical tasks, caring for patients, administrative tasks , write data, fill out forms and everything else. So, mitigating all of these things and keeping them out of the way gives medical experts more time to get the necessary information. And we need more time to get this information. So the only way to achieve this is to automate certain tasks in the process,” concludes Baresic.
