Predictive Models of Electronic Health Records

On Wednesday, March 10, ASEF Junior Fellow Primož Kocbek gave a lecture on electronic health records (EHR) as part of ASEF Young Mind popular science lectures. EHRs are health data of an individual or population that are collected over a long period of time and whose role is becoming increasingly important.
Primož emphasized at the lecture that there is a general need for high interpretability of predictive models in healthcare – both to improve trust on the part of users and to meet GDPR compliance, more precisely the patient’s right to interpret the results of the algorithm. However, there is no single methodology for assessing interpretability, so we may have different interpretable models with similar predictive accuracy but different interpretations.

The lecture, which was conducted as part of ASEF Young Minds, will soon be fully available on YouTube. SEF Junior Fellow Sara Košenina moderated the event. Read more about the lecture in this graphic.