In the future, artificial intelligence will play an important role in medicine. In diagnosis, successful tests have already been done: for example, the computer can learn to categorize images with great precision according to whether or not they show pathological changes. However, it is more difficult to train an artificial intelligence to examine the changing conditions of patients and calculate treatment suggestions; this is precisely what has now been achieved at TU Wien in cooperation with the Medical University of Vienna.
With the help of extensive data from intensive care units of various hospitals, an artificial intelligence has been developed that provides suggestions for the treatment of people who require intensive care due to sepsis. Analysis shows that artificial intelligence already exceeds the quality of human decisions. However, it is now important to also discuss the legal aspects of such methods.
Make optimal use of existing data
“In an intensive care unit, a lot of different data is collected throughout the day. Patients are constantly medically monitored. We wanted to investigate whether this data could be put to even better use than before,” says Professor Clemens Heitzinger from the Institute for Scientific Computing and Analysis at TU Wien (Vienna). He is also co-director of the “Center for Artificial Intelligence and Machine Learning” (CAIML) at the TU Vienna.
Medical personnel make their decisions based on well-founded rules. Most of the time, they know very well what parameters to take into account to provide the best care. However, the computer can easily take many more parameters into account than a human, and in some cases this can lead to better decisions.
The computer as planning agent
“In our project, we use a form of machine learning called reinforcement learning,” says Clemens Heitzinger. “It’s not just a simple categorization, for example, separating a large number of images into those that show a tumor and those that don’t, but a progression that changes over time, about the development that a given patient is likely to experience. Mathematically, this is quite a different thing. There has been little research on it in the medical field.”
The computer becomes an agent that makes its own decisions: if the patient is well, the computer is “rewarded.” If the condition deteriorates or death occurs, the computer is “grounded.” The computer program is tasked with maximizing its virtual “reward” by taking actions. In this way, extensive medical data can be used to automatically determine a strategy that achieves a particularly high probability of success.
already better than a human
“Sepsis is one of the most common causes of death in intensive care medicine and represents a huge challenge for clinicians and hospitals, as early detection and treatment are crucial for patient survival,” says Professor Oliver. Kimberger of the Medical University of Vienna. “Until now, there have been few medical advances in this field, making the search for new treatments and approaches all the more urgent. For this reason, it is especially interesting to investigate to what extent artificial intelligence can contribute to improving medical care. here. The use of machine learning models and other artificial intelligence technologies is an opportunity to improve the diagnosis and treatment of sepsis and ultimately increase the patient’s chances of survival.”
The analysis shows that AI capabilities are already outpacing humans: “Cure rates are now higher with an AI strategy than with purely human decisions. In one of our studies, the cure rate in terms of mortality at 90 days it increased by about 3% to about 88%,” says Clemens Heitzinger.
Of course, this does not mean that one should leave medical decisions in an intensive care unit to the computer alone. But AI can work as an additional device at the bedside, and medical staff can refer to it and compare their own assessment with the AI’s suggestions. These artificial intelligences can also be very useful in education.
Discussion on legal issues is necessary.
“However, this raises important questions, especially legal ones,” says Clemens Heitzinger. “Probably one first thinks about the question of who will be responsible for the mistakes made by artificial intelligence. But there is also the reverse problem: what if artificial intelligence had made the right decision, but the human chose a different treatment option and Did the patient suffer harm as a result?” Is the doctor then faced with the charge that he would have been better off relying on artificial intelligence because it comes with such a wealth of experience? Or should it be the right of the human being to ignore the advice of the computer at all times?
“The research project shows that artificial intelligence can already be used successfully in clinical practice with current technology, but a discussion about the social framework and clear legal rules are still urgently needed,” is convinced Clemens Heitzinger.
—————————————————-
Source link