“Segmenting Potentially Cancerous Areas in Prostate Biopsies using Semi-Automatically Annotated Data” by Nikolay Burlutskiy.
According to the MIDL 2019 Conference Program Chairs:
This paper presents an interesting and novel idea of a deep learning (DL) framework to segment potentially cancerous areas in prostate biopsies using semi-automatically generated data. All reviewers agreed on the significance, quality, clarity and alignment with MIDL conference. The application of this paper is quite innovative and the proposed method seems to be a direct assistance to the clinical routine.
See https://openreview.net/forum?id=SJxVA7xleE for full paper.