Vice-Dean Radek Martinek spoke at the Artificial Intelligence in Healthcare expert session held in the Chamber of Deputies of the Parliament of the Czech Republic. In his contribution entitled Artificial Intelligence and Innovative Technologies in Healthcare - Universities as a Key Partner for Politicians, Companies and Healthcare Facilities, he drew attention to the need for a responsible approach when deploying AI in clinical practice.
“In the dynamically evolving healthcare environment, it is imperative to ensure that the implementation of AI solutions is done with expert interpretation, meaningful interpretation and a focus on patient safety,” said Professor Martinek, who explained in his presentation that AI can be accurate but often does not understand the context as well as humans. It can therefore arrive at a result based on the wrong data context.
"That's why it's crucial to use explainable AI that gives doctors insight into the decision-making mechanisms of the model. This is the only way to ensure that AI makes decisions based on truly relevant information and that its outputs are trustworthy to physicians," Martinek said. In his presentation, he also presented the results of the Signal Lab research team, which he leads at the faculty. The team has long been involved in the development of advanced signal processing methods, smart sensing systems and bio-inspired algorithms for the healthcare industry. One of the main research directions is specifically the explainability of models in automatic classification and segmentation of MRI data. The goal is to ensure that artificial intelligence makes decisions based on actual pathological phenomena and not on random correlations or technical inaccuracies.
In conclusion, Vice Dean Martinek stressed that universities and research centres should be a partner in the development of legislation and strategic decisions in the field of digitalisation of healthcare.
“Without the involvement of experts, the wrong deployment of technologies can occur, which will ultimately endanger patients and public trust,” warned Radek Martinek, who also serves as Vice President for Implementation at the Czech Society for Artificial Intelligence and Innovative Digital Technologies in Healthcare (CSAIM). This professional platform brings together physicians, engineers, developers and researchers who work together to bring AI into everyday clinical practice in a safe and meaningful way.