Page 181 - IJB-8-1

RESEARCH ARTICLE

Analyzing Cell-Scaffold Interaction through

Unsupervised 3D Nuclei Segmentation

Kai Yao , Jie Sun *, Kaizhu Huang *, Linzhi Jing , Hang Liu , Dejian Huang , Curran Jude 2
1
1
3,4
1,2
4
3
1 School of Advanced Technology, Xi’an Jiaotong-Liverpool University, 111 Ren’ai Road, Suzhou, Jiangsu 215123, China
2 School of Engineering, University of Liverpool, The Quadrangle, Brownlow Hill, L69 3GH, UK
3 National University of Singapore (Suzhou) Research Institute, 377 Linquan Street, Suzhou, Jiangsu 215123, China
4 Department of Food Science and Technology, National University of Singapore, 3 Science Drive 2, 117542, Singapore
Abstract: Fibrous scaffolds have been extensively used in three-dimensional (3D) cell culture systems to establish in vitro
models in cell biology, tissue engineering, and drug screening. It is a common practice to characterize cell behaviors on such
scaffolds using confocal laser scanning microscopy (CLSM). As a noninvasive technology, CLSM images can be utilized
to describe cell-scaffold interaction under varied morphological features, biomaterial composition, and internal structure.
Unfortunately, such information has not been fully translated and delivered to researchers due to the lack of effective cell
segmentation methods. We developed herein an end-to-end model called Aligned Disentangled Generative Adversarial
Network (AD-GAN) for 3D unsupervised nuclei segmentation of CLSM images. AD-GAN utilizes representation
disentanglement to separate content representation (the underlying nuclei spatial structure) from style representation (the
rendering of the structure) and align the disentangled content in the latent space. The CLSM images collected from fibrous
scaffold-based culturing A549, 3T3, and HeLa cells were utilized for nuclei segmentation study. Compared with existing
commercial methods such as Squassh and CellProfiler, our AD-GAN can effectively and efficiently distinguish nuclei with
the preserved shape and location information. Building on such information, we can rapidly screen cell-scaffold interaction in
terms of adhesion, migration and proliferation, so as to improve scaffold design.
Keywords: Unsupervised learning; 3D nuclei segmentation; Aligned disentangled generative adversarial network; Fibrous
scaffold-based cell culture; Cell-scaffold interaction

*Correspondences to: Jie Sun, School of Engineering, University of Liverpool, The Quadrangle, Brownlow Hill, L69 3GH, UK; Jie.Sun@xjtlu.edu.
cn: Kaizhu Huang, School of Engineering, University of Liverpool, The Quadrangle, Brownlow Hill, L69 3GH, UK; Kaizhu.Huang@xjtlu.edu.cn

Received: October 25, 2021; Accepted: December 07, 2021; Published Online: December 30, 2021
Citation: Yao K, Sun J, Huang K, et al. 2022, Analyzing Cell-Scaffold Interaction through Unsupervised 3D Nuclei Segmentation. Int J
Bioprint, 8(1):495. http:// doi.org/10.18063/ijb.v8i1.495

1. Introduction these scaffolds using confocal laser scanning microscopy
(CLSM) . This technology scans the whole models
[3]
Scaffold-based three-dimensional (3D) cell culture layer by layer to collect CLSM image volumes and
systems have gained great attention as a replacement then stack them together. As a noninvasive technology,
of two-dimensional (2D) planar culture to mimic CLSM images can not only visualize cell behaviors,
extracellular matrix environments . Unlike the 2D but also reveal cell-scaffold interaction under varied
[1]
environment, 3D cell culture makes cell-cell contacts morphological features, biomaterial composition and
in all dimensions to obtain oxygen and nutrition, all of architectural structure. However, such information has
which lead to more in vivo-like gene expression and cell not been fully translated and delivered to researchers, due
behavior. Fibrous scaffolds have been extensively used to the complex nature of these images and the lack of
in 3D cell culture systems to establish in vitro models in effective analysis tools.
cell biology, tissue engineering, and drug screening . It To quantitatively analyze the cell culture model in
[2]
is a common practice to characterize cell behaviors on image-based cellular research, the first step is to extract
© 2021 Author(s). This is an Open Access article distributed under the terms of the Creative Commons Attribution License, permitting distribution and
reproduction in any medium, provided the original work is properly cited.
167