Thursday June 17 at 10am PT
Hosts: Moshe Safran (RSIP Vision) and Rabeeh Fares (RSIP Vision)
Invited speaker: Prof. Dr. Lena Maier-Hein, Head of Department, Computer Assisted Medical Interventions, DKFZ German Cancer Research Center.
Recent advances in deep learning have eliminated the need for handcrafted features carefully designed based on prior knowledge of a problem. Given these one-size-fits-all approaches as well as the increasing availability of large public data sets, one important question arises: Do we still need domain experts to solve medical image analysis problems?
Our speaker’s answer is Yes! Dr. Maier-Hein’s talk will highlight the importance of domain knowledge for various steps within the development process – from the selection of training/test data in the presence of possible confounders to the choice of appropriate validation metrics and the interpretation of algorithm results.
She will further present concepts for uncertainty handling in biomedical image analysis in which both novel machine learning-based methods, as well as traditional statistical methods, play a key role.