Uncertainty in Cardiovascular Modelling

Karolina Tlałka Represents Sano at UNCECOMP 2025

For a few days in June, the island of Rhodes became a hub for global discussion on uncertainty in modelling complex physical systems. The occasion was the 6th edition of UNCECOMP – International Conference on Uncertainty Quantification in Computational Science and Engineering (15-18 June 2025), which brought together researchers from around the world to share insights, results, and methods. Among the speakers was Karolina Tlałka, a PhD student in the Extreme-scale Data Computing team at Sano Centre for Computational Medicine. 

Karolina presented a talk titled “Some challenges associated with the personalisation and reduction of uncertainty in human cardiovascular models,” co-authored with Harry Saxton, Ian Halliday, Andrew Narracott, and Maciej Malawski. 

Her presentation addressed key challenges in the development of personalised cardiovascular models, with a particular focus on managing uncertainties in data and simulation methods. Such models are central to the advancement of personalised medicine, but their effectiveness relies on accurately representing patient-specific characteristics and understanding how uncertainty propagates through different stages of analysis. The authors show how to apply various types of boundaries on optimization process, to preserve physiological relevance of dynamic cardiovascular models. 

Attending UNCECOMP offered an excellent opportunity to exchange ideas with the international community of experts in computational mechanics, stochastic modelling, and multiscale analysis. The conference provided not only a platform to present research but also a space for initiating collaborations between research groups with complementary expertise.