Paper 12033-66
Paper 12033-66
Self-supervised U-Net for segmenting flat and sessile polyps
In person: 24 February 2022 • 11:50 AM - 12:10 PM PST 
Abstract
We propose a self-supervised approach to segment flat and sessile polyps. These polyps are particularly difficult to segment and are one of the main reasons for misdetection of polyps. We pre-train our self-supervised U-Net on Kvasir-SEG dataset followed by supervised training on the small Kvasir-Sessile dataset. We compare our self-supervised U-Net against fully supervised U-Net, Attention U-Net, R2U-Net, R2AU-Net and ResUNet++. We report an increase of dice coefficient by 0.29, 0.31, 0.32, 0.36 and 0.30, precision by 0.31, 0.39, 0.35, 0.36 and 0.35, and recall by 0.21, 0.18, 0.28, 0.35 and 0.2 due to self-supervision.
Presenter
Debayan Bhattacharya
Technische Univ. Hamburg-Harburg (Germany)
Debayan Bhattacharya is a PhD Student at Hamburg University of Technology and University Klinikum Hamburg-Eppendorf. He started his PhD in April of 2021. His research focuses on making machine learning models more accurate with limited ground truth. He also researchers in making machine learning models more explainable. When he is not researching, you will find him playing an electric guitar and jamming to the music of Beatles.