The AI4HF consortium presents a new episode of its podcast, released as part of the Cardio Talk: Latest Research series.
In this episode, Professor Kazem Rahimi (University of Oxford) explores how to leverage large-scale clinical data — such as electronic health records, biobanks, or hospital registries — to develop AI-based predictive models with the potential to improve cardiovascular disease (CVD) prevention and patient outcomes.
As Professor Rahimi explains:
“There are several research projects that make use of large clinical and research datasets for better prediction of heart failure, incidents of heart failure, or worsening of heart failure in people with pre-existing heart failure.”
This approach, he notes, also reinforces the relevance of existing methodologies:
“The calibration techniques we used for older cardiovascular models like Framingham are still valid and essential in this new context of high-dimensional AI modelling.”
Challenges such as data harmonisation, privacy, and applicability of models in different populations are also addressed.
The episode is available on ESC 365, Spotify, Apple Podcasts, and other platforms.
Click here to listen to the AI4HF podcast
The AI4HF (Artificial Intelligence for Heart Failure) project aims to harness the power of artificial intelligence to improve the prediction, prevention, and management of heart failure through the use of large-scale clinical data.
Click here to learn more about the AI4HF project