Deep learning based domain adaptation for mitochondria segmentation on EM volumes.

Journal: Computer methods and programs in biomedicine
Published Date:

Abstract

BACKGROUND AND OBJECTIVE: Accurate segmentation of electron microscopy (EM) volumes of the brain is essential to characterize neuronal structures at a cell or organelle level. While supervised deep learning methods have led to major breakthroughs in that direction during the past years, they usually require large amounts of annotated data to be trained, and perform poorly on other data acquired under similar experimental and imaging conditions. This is a problem known as domain adaptation, since models that learned from a sample distribution (or source domain) struggle to maintain their performance on samples extracted from a different distribution or target domain. In this work, we address the complex case of deep learning based domain adaptation for mitochondria segmentation across EM datasets from different tissues and species.

Authors

  • Daniel Franco-Barranco
    Donostia International Physics Center (DIPC), Donostia-San Sebastián, Spain. daniel_franco001@ehu.eus.
  • Julio Pastor-Tronch
    Dept. of Computer Science and Artificial Intelligence, University of the Basque Country (UPV/EHU), Spain.
  • Aitor González-Marfil
    Dept. of Computer Science and Artificial Intelligence, University of the Basque Country (UPV/EHU), Spain.
  • Arrate Muñoz-Barrutia
    Bioengineering and Aerospace Engineering Department, Universidad Carlos III de Madrid and Instituto de Investigación Sanitaria Gregorio Marañón, Madrid, Spain. mamunozb@ing.uc3m.es.
  • Ignacio Arganda-Carreras
    Ikerbasque, Basque Foundation for Science, Bilbao 48013, Spain.