Objectives and Scientific Topics
Data scarcity is arguably the single most prohibitive barrier to progress in medical image computing. Ostensibly, there are two ways to address it, (1) to collect, curate, and annotate massive datasets, and (2) to develop methods which achieve the same performance with less data. In its 4-year history, LABELS has convened one per year at the MICCAI conference to facilitate the presentation and discussion of ideas in both of these areas.
Researchers in this field are invited to submit abstracts and full papers for publication in the proceedings of each LABELS meeting. Accepted research is awarded a place in the poster session, and a select number of submissions are awarded 15 minute oral presentations.
In addition to the presentation of research, two to three invited keynote speakers provide their perspective on this important and rapidly evolving field. Further, one or more interactive sessions are held with the goal of producing a publication outlining open problems and recommendations for best practices in this area.