Take a vote! Help us enhance the interpretability of A.I. systems in Radiology:
Artificial Intelligence (A.I.) systems are achieving remarkable performances in radiology, but at the cost of increased complexity. Hence, they become less interpretable.
As these systems are pervasively being introduced to radiology, we believe it becomes imperative to develop dedicated methodologies to enhance the interpretability of A.I. technologies.
Interpretability methods could help physicians to decide whether they should follow/trust a prediction from an A.I. system. Ultimately, interpretability of A.I. systems is closely linked to safety in healthcare.
The following poll is meant to collect radiologists’ opinions about methods to enhance the interpretability of A.I. systems developed to assist radiologists.
https://www.surveymonkey.com/r/interpretableAI
We thank you in advance for taking 5 minutes to answer this poll. The results of the poll are going to be made publicly available and part of a related publication where this topic will be discussed.
On behalf of the organisers and supporters of iMIMIC Workshop (Interpretability of Machine Intelligence in Medical Image Computing)