Self-mentoring: A new deep learning pipeline to train a self-supervised U-net for few-shot learning of bio-artificial capsule segmentation.

Journal: Computers in biology and medicine
Published Date:

Abstract

BACKGROUND: Accurate segmentation of microscopic structures such as bio-artificial capsules in microscopy imaging is a prerequisite to the computer-aided understanding of important biomechanical phenomenons. State-of-the-art segmentation performances are achieved by deep neural networks and related data-driven approaches. Training these networks from only a few annotated examples is challenging while producing manually annotated images that provide supervision is tedious.

Authors

  • Arnaud Deleruyelle
    University Lille, CNRS, Centrale Lille, UMR 9189 - CRIStAL, F-59000 Lille, France. Electronic address: arnaud.deleruyelle@univ-lille.fr.
  • Cristian Versari
    University Lille, CNRS, Centrale Lille, UMR 9189 - CRIStAL, F-59000 Lille, France. Electronic address: cristian.versari@univ-lille.fr.
  • John Klein
    University Lille, CNRS, Centrale Lille, UMR 9189 - CRIStAL, F-59000 Lille, France. Electronic address: john.klein@univ-lille.fr.