CASTLE: cell adhesion with supervised training and learning environment

Research output: Contribution to journalArticle

Standard

CASTLE: cell adhesion with supervised training and learning environment. / Gilbert, S G; Krautter, F; Cooper, D; Chimen, M; Iqbal, A J; Spill, F.

In: Journal of Physics D: Applied Physics, Vol. 53, No. 42, 424002, 14.10.2020.

Research output: Contribution to journalArticle

Harvard

APA

Vancouver

Author

Bibtex

@article{ecfe1f03d41d4ef491525677e6df34b9,
title = "CASTLE: cell adhesion with supervised training and learning environment",
abstract = "Different types of microscopy are used to uncover signatures of cell adhesion and mechanics. Automating the identification and analysis often involve sacrificial routines of cell manipulation such as in vitro staining. Phase-contrast microscopy (PCM) is rarely used in automation due to the difficulties with poor quality images. However, it is the least intrusive method to provide insights into the dynamics of cells, where other types of microscopy are too destructive to monitor. In this study, we propose an efficient workflow to automate cell counting and morphology in PCM images. We introduce Cell Adhesion with Supervised Training and Learning Environment (CASTLE), available as a series of additional plugins to ImageJ. CASTLE combines effective techniques for phase-contrast image processing with statistical analysis and machine learning algorithms to interpret the results. The proposed workflow was validated by comparing the results to a manual count and manual segmentation of cells in images investigating different adherent cell types, including monocytes, neutrophils and platelets. In addition, the effect of different molecules on cell adhesion was characterised using CASTLE. For example, we demonstate that Galectin-9 leads to differences in adhesion of leukocytes. CASTLE also provides information using machine learning techniques, namely principal component analysis and k-means clustering, to distinguish morphology currently inaccessible with manual methods. All scripts and documentation are open-source and available at the corresponding GitLab project.",
keywords = "cell adhesion, cell morphology, galectins, ilastik, image analysis, imagej, machine learning",
author = "Gilbert, {S G} and F Krautter and D Cooper and M Chimen and Iqbal, {A J} and F Spill",
year = "2020",
month = oct,
day = "14",
doi = "10.1088/1361-6463/ab9e35",
language = "English",
volume = "53",
journal = "Journal of Physics D: Applied Physics",
issn = "0022-3727",
publisher = "IOP Publishing",
number = "42",

}

RIS

TY - JOUR

T1 - CASTLE: cell adhesion with supervised training and learning environment

AU - Gilbert, S G

AU - Krautter, F

AU - Cooper, D

AU - Chimen, M

AU - Iqbal, A J

AU - Spill, F

PY - 2020/10/14

Y1 - 2020/10/14

N2 - Different types of microscopy are used to uncover signatures of cell adhesion and mechanics. Automating the identification and analysis often involve sacrificial routines of cell manipulation such as in vitro staining. Phase-contrast microscopy (PCM) is rarely used in automation due to the difficulties with poor quality images. However, it is the least intrusive method to provide insights into the dynamics of cells, where other types of microscopy are too destructive to monitor. In this study, we propose an efficient workflow to automate cell counting and morphology in PCM images. We introduce Cell Adhesion with Supervised Training and Learning Environment (CASTLE), available as a series of additional plugins to ImageJ. CASTLE combines effective techniques for phase-contrast image processing with statistical analysis and machine learning algorithms to interpret the results. The proposed workflow was validated by comparing the results to a manual count and manual segmentation of cells in images investigating different adherent cell types, including monocytes, neutrophils and platelets. In addition, the effect of different molecules on cell adhesion was characterised using CASTLE. For example, we demonstate that Galectin-9 leads to differences in adhesion of leukocytes. CASTLE also provides information using machine learning techniques, namely principal component analysis and k-means clustering, to distinguish morphology currently inaccessible with manual methods. All scripts and documentation are open-source and available at the corresponding GitLab project.

AB - Different types of microscopy are used to uncover signatures of cell adhesion and mechanics. Automating the identification and analysis often involve sacrificial routines of cell manipulation such as in vitro staining. Phase-contrast microscopy (PCM) is rarely used in automation due to the difficulties with poor quality images. However, it is the least intrusive method to provide insights into the dynamics of cells, where other types of microscopy are too destructive to monitor. In this study, we propose an efficient workflow to automate cell counting and morphology in PCM images. We introduce Cell Adhesion with Supervised Training and Learning Environment (CASTLE), available as a series of additional plugins to ImageJ. CASTLE combines effective techniques for phase-contrast image processing with statistical analysis and machine learning algorithms to interpret the results. The proposed workflow was validated by comparing the results to a manual count and manual segmentation of cells in images investigating different adherent cell types, including monocytes, neutrophils and platelets. In addition, the effect of different molecules on cell adhesion was characterised using CASTLE. For example, we demonstate that Galectin-9 leads to differences in adhesion of leukocytes. CASTLE also provides information using machine learning techniques, namely principal component analysis and k-means clustering, to distinguish morphology currently inaccessible with manual methods. All scripts and documentation are open-source and available at the corresponding GitLab project.

KW - cell adhesion

KW - cell morphology

KW - galectins

KW - ilastik

KW - image analysis

KW - imagej

KW - machine learning

UR - http://www.scopus.com/inward/record.url?scp=85090013315&partnerID=8YFLogxK

U2 - 10.1088/1361-6463/ab9e35

DO - 10.1088/1361-6463/ab9e35

M3 - Article

VL - 53

JO - Journal of Physics D: Applied Physics

JF - Journal of Physics D: Applied Physics

SN - 0022-3727

IS - 42

M1 - 424002

ER -