AI holds a promise to revolutionize healthcare. It has the potential to improve the quality of care by improving diagnosis and prognosis, reducing growing healthcare costs through automation, and making care widely available. However, contrary to the general impression that AI is widely utilized in hospitals, the reality is a bit more nuanced.
There are still questions to be addressed regarding the ability of the methods to generalize and adapt to new data, interpretability of the automatic decisions, the ability of the methods to automatically control the quality of their performance.
Through the research performed at the Quantitative healthcare analysis group (qurAI), which is embedded in the Faculty of Medicine and the Faculty of Science at the University of Amsterdam, professor Ivana Išgum illustrates how we develop and evaluate novel AI solutions to solve data analysis challenges encountered in different steps of the patient pathway and challenges towards clinical integration of the AI methods.