Imaging data analysis using non-negative matrix factorization

Toru Aonishi; Ryoichi Maruyama; Tsubasa Ito; Hiroyoshi Miyakawa; Masanori Murayama; Keisuke Ota

doi:10.1016/j.neures.2021.12.001

Imaging data analysis using non-negative matrix factorization

Neurosci Res. 2022 Jun:179:51-56. doi: 10.1016/j.neures.2021.12.001. Epub 2021 Dec 22.

Authors

Toru Aonishi¹, Ryoichi Maruyama², Tsubasa Ito³, Hiroyoshi Miyakawa⁴, Masanori Murayama⁵, Keisuke Ota⁶

Affiliations

¹ School of Computing, Tokyo Institute of Technology, Kanagawa, Japan; Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology, Kanagawa, Japan; RIKEN Center for Brain Science, Saitama, Japan. Electronic address: aonishi@c.titech.ac.jp.
² Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology, Kanagawa, Japan.
³ Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology, Kanagawa, Japan; RIKEN Center for Brain Science, Saitama, Japan.
⁴ School of Life Sciences, Tokyo University of Pharmacy and Life Sciences, Tokyo, Japan.
⁵ RIKEN Center for Brain Science, Saitama, Japan.
⁶ RIKEN Center for Brain Science, Saitama, Japan; Graduate School of Medicine, The University of Tokyo, Tokyo, Japan.

PMID: 34953961
DOI: 10.1016/j.neures.2021.12.001

Abstract

The rapid progress of imaging devices such as two-photon microscopes has made it possible to measure the activity of thousands to tens of thousands of cells at single-cell resolution in a wide field of view (FOV) data. However, it is not possible to manually identify thousands of cells in such wide FOV data. Several research groups have developed machine learning methods for automatically detecting cells from wide FOV data. Many of the recently proposed methods using dynamic activity information rather than static morphological information are based on non-negative matrix factorization (NMF). In this review, we outline cell-detection methods related to NMF. For the purpose of raising issues on NMF cell detection, we introduce our current development of a non-NMF method that is capable of detecting about 17,000 cells in ultra-wide FOV data.

Keywords: Cell detection; Machine learning; Multicellular calcium imaging; Region of interest; Wide field-of-view microscope.

Publication types

Review

MeSH terms

Algorithms*
Data Analysis*
Diagnostic Imaging
Machine Learning