Artificial Intelligence News

Artificial intelligence in the intensive care unit


In the future, artificial intelligence will play an important role in medicine. In diagnostics, successful tests have already been performed: for example, computers can learn to categorize images with great accuracy based on whether or not they show pathological changes. However, it is more difficult to train artificial intelligence to examine a patient’s condition over time and to calculate treatment suggestions – this is what has now been achieved at TU Wien in collaboration with the Medical University of Vienna.

In the future, artificial intelligence will play an important role in medicine. In diagnostics, successful tests have already been performed: for example, computers can learn to categorize images with great accuracy based on whether or not they show pathological changes. However, it is more difficult to train artificial intelligence to examine a patient’s condition over time and to calculate treatment suggestions – this is what has now been achieved at TU Wien in collaboration with the Medical University of Vienna.

With the help of extensive data from the intensive care units of various hospitals, artificial intelligence was developed that provides advice for the treatment of people requiring intensive care due to sepsis. Analysis shows that artificial intelligence has already surpassed the quality of human decisions. However, it is also important now to discuss the legal aspects of the method.

Make optimal use of existing data

“In the intensive care unit, a lot of different data is collected all the time. The patient continues to be medically monitored. We want to investigate whether this data can be used better than before,” said Prof. Clemens Heitzinger of the Institute for Analysis and Scientific Computing at TU Wien (Vienna). He is also the cross-faculty Co-Director of the “Center for Artificial Intelligence and Machine Learning” (CAIML) at TU Wien.

Medical staff make their decisions based on reasoned rules. Most of the time, they know very well which parameters they have to take into account in order to provide the best treatment. However, computers can easily take many more parameters into account than humans – and in some cases this can lead to better decisions.

The computer as a planning agent

“In our project we use a form of machine learning called reinforcement learning,” says Clemens Heitzinger. “It’s not just about simple categorization – for example, separating a large number of images into those that show tumors and those that don’t – but about developments that change transiently, about developments that a particular patient might have. Mathematically, this is something completely different. There is little research into this in the medical field.”

The computer becomes the agent making its own decisions: if the patient is doing well, the computer is “rewarded”. If conditions worsen or death occurs, the computer is “punished”. Computer programs have a job to maximize their virtual “reward” by taking action. In this way, extensive medical data can be used to automatically define strategies that achieve a very high probability of success.

It’s better than humans

“Sepsis is one of the most common causes of death in intensive care medicine and poses a big challenge for doctors and hospitals, because early detection and treatment is very important for patient survival,” said Prof. Oliver Kimberger of the Medical University of Vienna. “So far, there have been few medical breakthroughs in this area, which makes the search for new treatments and approaches even more urgent. For this reason, it is very interesting to investigate to what extent artificial intelligence can contribute to improving medical care here. Using machine learning models and other AI technologies is an opportunity to improve the diagnosis and treatment of sepsis, ultimately increasing the chances of patient survival.”

The analysis shows that AI capabilities already outperform humans: “The cure rate is now higher with AI strategies than with purely human decisions. In one of our studies, the cure rate in terms of 90-day mortality increased by about 3% to about 88%,” said Clemens Heitzinger.

Of course, this doesn’t mean that one should leave medical decisions in the intensive care unit to the computer alone. But artificial intelligence can work as a bedside enhancement – ​​and medical staff can consult and compare their own judgments with artificial intelligence advice. Such artificial intelligence can also be very useful in education.

Discussion of legal issues is required

“However, this raises important questions, especially legal ones,” said Clemens Heitzinger. “One might ponder the question of who will be held responsible for any mistakes artificial intelligence makes first. But there is also the opposite problem: what if the artificial intelligence has made the right decision, but humans choose different treatment options and the patient suffers a loss as a result?” Do doctors then face accusations that it is better to trust artificial intelligence because it is equipped with a lot of experience? Or should it be a human right to ignore the computer’s suggestions every time?

“The research project shows: artificial intelligence can already be used successfully in clinical practice with today’s technology – but discussions about social frameworks and clear rules of law are still urgently needed,” believes Clemens Heitzinger.




Source link

Related Articles

Back to top button