Skip to content

Evaluating adipocyte differentiation of bone marrow-derived mesenchymal stem cells by a deep learning method for automatic lipid droplet counting

Notifications You must be signed in to change notification settings

hassanlougithub/lipid_droplets_counting

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 

Repository files navigation

Evaluating adipocyte differentiation of bone marrow-derived mesenchymal stem cells by a deep learning method for automatic lipid droplet counting

Link

Abstract

Stem cells are a group of competent cells capable of self-renewal and differentiating into osteogenic, chondrogenic, and adipogenic lineages. These cells provide the possibility of successfully treating patients. During differentiation into adipose tissues, a large number of lipid droplets normally accumulate in these cells, which can be seen through oil red O staining. Although the oil red O staining technique is regularly used for assessing the differentiation degree, its validity for quantitative studies has not been approved yet. Lipid droplet counting has applications in differentiation works and saves time and costs once being automated. In this research, for proving the differentiation of mesenchymal stem cells (MSCs) into adipocyte tissues, their microscopic images were provided. Then, the microscopic images were segmented into square patches, and the lipid droplets were annotated through single-point annotation. The proposed network, based on deep learning, is a fully convolutional regression network processing an image with a small respective field on it. Finally, this method not only does count the lipid droplets but also generates a count map. The average counting accuracy is 94%, which is higher than that of the state-of-the-art methods. It is useful to cell biologists to check the percentage of differentiation in different samples. Also, with a count map, it is possible to observe the regions with high concentrations of lipid droplets without oil red O staining and, thus, examine the total adipocyte differentiation. The contribution of this paper is that a deep learning algorithm has been used for the first time in the field of processing intracellular images.

image image

About

Evaluating adipocyte differentiation of bone marrow-derived mesenchymal stem cells by a deep learning method for automatic lipid droplet counting

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages