A Primer on Deep Learning-Based Cellular Image Classification of Changes in the Spatial Distribution of the Golgi Apparatus After Experimental Manipulation.

Journal: Methods in molecular biology (Clifton, N.J.)
PMID: 36512222

Abstract

The visual classification of cell images according to differences in the spatial patterns of subcellular structure is an important methodology in cell and developmental biology. Experimental perturbation of cell function can induce changes in the spatial distribution of organelles and their associated markers or labels. Here, we demonstrate how to achieve accurate, unbiased, high-throughput image classification using an artificial intelligence (AI) algorithm. We show that a convolutional neural network (CNN) algorithm can classify distinct patterns of Golgi images after drug or siRNA treatments, and we review our methods from cell preparation to image acquisition and CNN analysis.

Authors

  • Daisuke Takao
    Department of Cell Biology and Anatomy and International Research Center for Neurointelligence (WPI-IRCN), Graduate School of Medicine, The University of Tokyo, Tokyo 113-0033, Japan.
  • Yuki M Kyunai
    Faculty of Engineering, Department of Applied Chemistry and Biotechnology, Okayama University, Okayama, Japan.
  • Yasushi Okada
    Department of Cell Biology and Anatomy and International Research Center for Neurointelligence (WPI-IRCN), Graduate School of Medicine, The University of Tokyo, Tokyo 113-0033, Japan.
  • Ayano Satoh
    Graduate School of Interdisciplinary Science and Engineering in Health Systems, Okayama University, Okayama, Japan. ayano113@cc.okayama-u.ac.jp.

Keywords

External Resources

Popular Topics
Recent Journals