AI | Algit d.o.o.

Exemplary Case of Work Optimization with AI: Idrija Psychiatric Hospital

Vid Ogris — Mon, 17 Jul 2023 10:21:30 +0000

Modern healthcare faces numerous challenges that need to be addressed in order to improve the quality of patient care. One of the most significant challenges is the shortage of healthcare workers, particularly nurses. Slovenia is no exception. This problem seriously affects the efficiency and quality of the healthcare system. However, a revolutionary solution is on the horizon – the use of artificial intelligence.

With the help of advanced artificial intelligence algorithms, it is possible to achieve better organization of nurses’ work schedules, significantly reducing overload and improving employee satisfaction. Smart systems, such as Woshi, enable automated scheduling and adjustment of schedules based on employees’ needs. This helps reduce overtime, improve working conditions, and ensure quality patient care.

Good Practice in Slovenia

Fortunately, we have some excellent examples of hospitals in Slovenia that have successfully implemented such digital solutions. One such example is the psychiatric hospital in Idrija, where the use of artificial intelligence systems for organizing nurses’ work schedules was successfully implemented in just a few months.

“We recognized the need for an efficient and optimized scheduling process that would allow us to better utilize our staff. Prior to implementing the WoShi system, the process of creating schedules for our 110 employees required a significant amount of time and intellectual effort. On average, we spent about seven working days to create an optimal schedule, facing challenges in finding the most suitable solutions for each employee. With WoShi, we managed to reduce the time spent on scheduling to approximately one working day. WoShi has taken over the majority of the work and, after six months, it handles around 90% of the tasks that we previously did manually. This allows us to focus on other important tasks,” says Urban Bole, MSc in Nursing and Assistant Director for Nursing at Idrija Psychiatric Hospital. “In addition to automated scheduling, it has been shown that the most useful feature of WoShi for supervisors is the ability to quickly experiment with different ideas and suggestions that employees have or that the workflow itself requires. This functionality provides us with flexibility in adjusting the schedule according to changing needs and employee preferences. WoShi uses advanced algorithms to find solutions that are difficult for a person, regardless of their knowledge and experience, to achieve,” Bole added.

Systematic Implementation of Artificial Intelligence in the Healthcare Sector is Essential

To harness the benefits of artificial intelligence in healthcare, it is crucial for healthcare institutions’ leadership to take a leading role in the adoption and support of these technologies. It is important to raise awareness about the potential of artificial intelligence and its advantages while addressing any potential fears. In order to ensure their readiness for the use of new technologies, systematic training for nurses in working with smart systems must be implemented. Leadership must take responsibility for promoting changes that will contribute to a more efficient and advanced healthcare system.

WoShi has taken over the majority of the work and, after six months, it handles around 90% of the tasks that we previously did manually.

The ability to quickly experiment with different ideas and suggestions allows for greater flexibility in accommodating employee preferences.

Digitization in healthcare and why the delay

Vid Ogris — Thu, 08 Jun 2023 09:45:42 +0000

Estonia is often cited as an example for the reform of the public health system in Slovenia, especially due to digitalization. In Estonia, as many as 96% of the population have health records organized in the Digilugu.ee digital system, where all medical data, referrals, sick leave and preventive programs are transparently accessible.

The Ministry of Health of Slovenia is of the opinion that Estonia can be an example, but health reforms in our country will be undertaken according to the Slovenian model, which still lags behind the Estonian and Finnish models. Tjaša Zajc, an expert in the field of digital health employed by Better, emphasizes that unified standards, such as e.g. e-ordering.

Digitization is not only electronic files, but also digitization of processes and connectivity of systems. New solutions make it possible to optimize work of doctors and order patients, which reduces costs. Artificial intelligence can help process unstructured data, such as surveys, and transform them into structured data. “A lot of things in Slovenian hospitals, like in other countries, are still done on paper. We have a problem with the fragmentation of systems. Much work will need to be done that the doctors will not have to repeatedly enter the same data into different systems. If you had more things in the same place, you would be able to orient yourself better.”

Slovenia must unify the fragmented systems and introduce digital scheduling for greater efficiency. E-ordering (scheduling patients for appointments) in Slovenia, compared to Estonia or Finland, still does not work well. The system is still fragmented across different pages that patients must navigate on their own, and the accuracy of the data is questionable. The key challenge is the development of a high-quality and secure system within the country, as external providers are not an ideal solution. It is difficult for the government to compete with the private sector in paying IT professionals. “The idea of a state-owned company that would deal with the digitization of healthcare is interesting, but it would require sufficient funds and a strong staff.”

We need more investments.

In order to achieve greater uniformity and improve the user experience, more investment in digitization is needed. “Digitalization in Slovenia takes place unsystematically and with very little investment for hospitals. Our hospitals give one percent for computerization. Hospitals receive about 3 percent of the state budget allocated for digitization,” says Zajc, adding that more investment in digitization and the unification of standards would contribute to a better user experience for patients and medical staff.

The question is how quickly and successfully the digitization of healthcare will take place, as it depends on project management, policy and resources. Digital transformation involves changing processes and introducing new, optimized solutions. The debate on the digitization of healthcare will continue, as there will be no shortage of challenges in Slovenian healthcare and society.

Summary of the radio report Aktualno 202: How to digitize Slovenian healthcare (April 5, 2023). Author: Luka Hvalc

Evolutionary Computing vs. Convolutional Neural Networks

Kristl Ogris — Fri, 11 Aug 2017 00:49:51 +0000

There is a significant number of problems, which just too hard for classical algorithms. Even if we were able to harness all the computer power in the world, it wouldn’t help much.

For example, the problems in operational research, like scheduling, which tend to trigger the so called combinatoric explosion, are one example. Classification of images and speech, handwriting recognition, taxonomical problems and so on. In the last 50 years, we had several promising ideas about how to tackle those. After some time the initial excitement subsided due to the lack of results, and the AI winter came once again.

Neural Networks

But then, around 2012, Neural Networks came back once again. Maybe for good this time. They used to be too computationally expensive for the hardware we had. But now, after several decades of Moore’s law, this has become less of a problem. Besides, some important algorithmic breakthroughs have been also made.

Neural networks, convolutional or otherwise, came by and swept away all those much more orthodox solutions championed and financed by big companies like Microsoft, which has invested billions into solving speech recognition the old fashioned way. Neural networks were a black swan event and an extinction event for a great part of the AI efforts of that time. When Lee Sedol lost against AlphaGo, the AI skeptic community of naysayers was damaged as well. Suddenly Neural Networkshad the potential to do almost everything and nobody would be able to stop them from becoming superintelligent one day.

When a great enthusiasm overwhelms more people than just a few mad scientists, as is now the case, it’s a great time to be sober. Whatever is the highest fashion of them all, might not stay with us forever. NN’s are fashionable, all right, but they will probably never become a classic. We have to ask ourselves, where these Neural Networks have come from? Is there already something new on the horizon? The scientific foundation for neural networks is a 70 and more years old story. After that, we had some good decades of exponential growth in computing. Early in this century when people like Jurgen Schmidthuber completed everything with scientific papers and performed some experiments only a few took notice. It was the year 2009 when one ought to have placed his bet on Neural Networks which are already too fashionable today.

Evolutionary Computing

Evolutionary Computing, which is even older in the sense that Charles Darwin laid its foundations, was also introduced into computing long ago, by John Koza. The start was rough, even more than it was with Neural Nets. The scientific fundamentals were further improved in the 20th and 21st century, but no Nobel prize was awarded. Stockholm isn’t that great a visionary. Besides the fact, that EC is more general in principle than NN, there are several more important advantages of EC over NN:

EC can invent new stuff. While NN is based on some learning from human experts, EC can always start from scratch if need be. The creativity of EC might remind you of the creativity of biological evolution. This is no accident, but an artifact of its principles.
EC can use NN modules or any other algorithms when appropriate. Whereas NN isn’t ascompatible with everything.
EC is an example of a fully parallel algorithm. NN, not so much.
NN solutions are always opaque, EC solutions are crisp as you want them to be.
EC needs little data, but can handle a lot of it. SmallData is useless for NN.