AI is making remarkable progress in the healthcare industry and has potential applications in many aspects of the health industry, including clinical decision-making, public health, biomedical research, drug development, health systems management, and service redesign. there is.
While many countries around the world are early adopters of AI in this field, Croatia is at the forefront of the CEE region by leveraging AI to advance the healthcare environment.
Over the past few years, efforts in the country have been led by AI4HEALTH, a non-profit consortium of experts in the AI, healthcare, and startup fields. It all started when a consortium of researchers saw an opportunity for Croatia to set up a so-called European Digital Innovation Hub (EDIH) for the application of her AI technology.
“This was a great opportunity to create a hub that revolved around our AI background, recognizing how technology permeates healthcare and medicine. I think it’s very timely and very valuable to move healthcare forward faster,” said Anja Baresic. AI4Health coordinator tells her The Recursive.
Around the EU, these digital innovation hubs are entrepreneurs supporting businesses, the public sector and other stakeholders in the process of digitizing their business, developing digital skills and mediating between service providers and users in new technology areas. Designed as an infrastructure to support the home spirit. Areas of artificial intelligence, cybersecurity, and high performance computing (HPC).
Baresic himself has extensive experience in machine learning and computational biology and is a researcher at the Rudjer Boskovic Institute, the largest scientific research center in the country. Like other researchers and his AI4HEALTH members, Baresic first saw a challenge for Croatian healthcare, which was that although the technology was developing fairly quickly, it lacked application.
“There are two outliers that make it really difficult. Some people shy away from AI if they are doing anything in the field, so perhaps most of these companies have the same problems and need help solving these challenges faster. I figured there was,” explains Baresic.
What AI4HEALTH did next was start creating a digital healthcare ecosystem of sorts.
“We have brought together everyone in the Croatian digital healthcare sector. Both the regulatory side, the technology companies that provide the infrastructure, the academia, the teaching faculties, the incubation and the financial side. There are also several healthcare companies interested in being a proving ground for our services,” notes Baresic.
Practical application of AI in Croatian hospitals
Some of the AI solutions developed by the startup have already been adopted by Croatian medical institutions.
One use case is a speech-to-text solution developed by Croatian AI startup Newton Technologies Adria, developed specifically for the Croatian language and trained on medical records.
As Baresic explains, the solution is used in more than 200 general and hospital settings nationwide.
“What they’ve done is develop tailor speech-to-text for medical texts in local languages, which is obviously very difficult because these languages are so complex. It provides clinicians with the service of speaking into a dictation device, which is translated into text, making the process of filling out medical reports 8x to 10x faster with just a few checks and corrections. It’s been very useful and has been used quite a lot,” Baresic told The Recursive.
Another practical case is using wearable ECG (electrocardiogram) devices and AI solutions to improve the diagnosis and treatment of atrial fibrillation or arrhythmia. Combining Slovenian technology with an AI solution developed by Polish startup Caridomatics, the device could potentially record signals for up to seven days, whereas conventional signals can only record his 24 hours.
“The idea here is that if it can be detected and treated early, it has great preventive value. This setup is also more likely to be used in primary care because they don’t need to be referred to a pediatrician, so it gives them a nice little report that gives them more information. Handing over from next-practitioner to general practitioner means more accessibility. For example, in rural areas there is always the challenge of sending someone to the nearest cardiologist who is miles away,” Baresic said. told The Recursive.
Then there is also the use of conversational AI. Croatian startup Mindsmiths and its chatbot Megi, for example, are launching towards the end of his 2020, integrating into local hospitals to help chronically ill patients suffering from high blood pressure.
“Chatbots help track blood pressure and heart rate, collect data by entering it into the system, and publish those data to cardiologists on the other side. It can also lead to active participation, but this is probably the only way for chronic illness,” Baresic stresses.
Potential for further use of new AI solutions
Umo neuroscience is a Croatian startup developing neurofeedback and biofeedback technologies and processes. According to CEO and founder Marin Maletic, the use of AI in healthcare will reach new heights in the ensuing period.
“We clearly understand that in our business of developing and delivering innovative neurofeedback services, we are at the intersection of healthcare and wellbeing. , customer management, developing neurofeedback protocols, and using AI in the areas of EEG signal processing.Practically everywhere we look, we see AI as a potentially great tool for improvement. is.” Recursive.
But there is room for improvement, which he said could be achieved by adopting a more humane approach to AI, as some companies in Croatia are already doing.
“So many people feel alone and in the dark when they have a serious diagnosis. It causes pain and fear. The healthcare system inevitably treats humans like statistics and lacks the resources to make them agents of the healing process. I think they are a great example of a humanistic approach to AI that I generally like very much.
In that sense, according to Baresic, AI technology was never meant to replace doctors, just to help them do their jobs better and spend more quality time with their patients. . According to Deloitte’s report, AI applications could free him 1.944 billion hours from 1.659 billion each year.
“Studies show that you should always have a 20-minute consultation with a specialist. , physical work, patient care, administrative work, writing down data, filling out forms, and everything else. The house has more time to get the insights it needs, and more time needs to be devoted to those insights, so the only way to do that is to automate some of the tasks in the process,” Baresic says. concludes Mr.