Machines can learn to not only make predictions but also process causal relationships. An international research team is showing how this can make treatment safer, more efficient and more personalized.
Artificial intelligence is also advancing in the medical field. When it comes to imaging technology and health risk calculations, there are a number of AI methods in development and testing stages. Machines are expected to greatly benefit humanity when it comes to recognizing patterns in large amounts of data. Following the classical model, AI compares information to learned examples, draws conclusions, and makes inferences.
Currently, an international team led by Professor Stefan Feueriger, Director of the Institute for Artificial Intelligence (AI) Management at LMU, is exploring the potential of AI in a relatively new field for diagnosis and treatment. Can causal machine learning (ML) estimate treatment outcomes better than previously commonly used ML techniques? Yes, the group's groundbreaking research published in a prestigious journal Research says so. natural medicine: Causal ML can improve treatment efficacy and safety.
In particular, new machine learning variants offer “rich opportunities to individually improve patient health by personalizing treatment strategies,” say researchers from Munich, Cambridge (UK) and Boston (US). is writing. He is Professor of Computer Science at the Technical University of Munich (TUM) and the leader of the Helmholtz AI group, which includes Stefan Bauer and Niki Kilbertus.
Regarding machine assistance in treatment decisions, the authors expect decisive advances in quality. They claim that classical ML recognizes patterns and discovers correlations. However, the causal principle of cause and effect remains, in principle, closed to machines. They are unable to address questions about why. However, many of the questions that arise when making treatment decisions involve issues of causation. The authors illustrate this with the example of diabetes. Classical ML aims to predict the likelihood of disease in a given patient with various risk factors. Causal ML ideally makes it possible to answer how risks arise. change If the patient was given antidiabetic drugs. That is, evaluate the effect of the cause (medication prescription). For example, it is possible to estimate whether another treatment plan is better than the commonly prescribed drug metformin.
However, to be able to estimate the effects of a hypothetical treatment, “the AI model needs to learn how to answer the 'what if?' questions. Naturally,” says a doctoral candidate on Feurigel's team. says Jonas Schweithal. “We give the machine rules to recognize the causal structure and formalize the problem correctly,” Feurigel says. Next, the machine must learn to recognize the effects of the intervention and, so to speak, understand how real-world results are reflected in the data entered into the computer.
Even in situations where reliable treatment standards do not yet exist, or where randomized studies are not possible for ethical reasons because they always include a placebo group, the machine can assess potential treatment outcomes from available patient data and determine what is possible. You can formulate a hypothesis for a treatment plan. , the researchers hope. The use of such real-world data should generally allow estimates to describe patient cohorts more accurately than ever before, thereby allowing more accurate individualized treatment decisions. It will be. Naturally, challenges still exist to ensure the reliability and robustness of the method.
The software required for causal ML techniques in healthcare does not exist out of the box. ”
Professor Stefan Feuerriegel, Director of the LMU Institute of Artificial Intelligence (AI) Management
Rather, it requires “complex modeling” of each problem, with “close collaboration between AI experts and doctors.” Like his TUM colleagues Stefan Bauer and his Niki Kilbertus, Feuerriegel is also a professor at the Munich Center for Machine Learning (MCML) and the Konrad Zuse School of Excellence in Reliable AI in AI in healthcare, decision-making and other topics. researching issues related to. In other application areas, such as marketing, causal ML research has already been in the testing phase for several years, Feurigel explains. “Our goal is to bring this technique one step closer to practice. This paper describes the direction in which things could move in the coming years.”
sauce:
Ludwig Maximilian University of Munich (LMU)
