Environmental properties of cells improve machine learning-based phenotype recognition accuracy

Timea Toth, Tamas Balassa, Norbert Bara, Ferenc Kovacs, Andras Kriston, Csaba Molnar, L. Haracska, Farkas Sukosd, Peter Horvath

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

To answer major questions of cell biology, it is often essential to understand the complex phenotypic composition of cellular systems precisely. Modern automated microscopes produce vast amounts of images routinely, making manual analysis nearly impossible. Due to their efficiency, machine learning-based analysis software have become essential tools to perform single-cell-level phenotypic analysis of large imaging datasets. However, an important limitation of such methods is that they do not use the information gained from the cellular micro- and macroenvironment: the algorithmic decision is based solely on the local properties of the cell of interest. Here, we present how various features from the surrounding environment contribute to identifying a cell and how such additional information can improve single-cell-level phenotypic image analysis. The proposed methodology was tested for different sizes of Euclidean and nearest neighbour-based cellular environments both on tissue sections and cell cultures. Our experimental data verify that the surrounding area of a cell largely determines its entity. This effect was found to be especially strong for established tissues, while it was somewhat weaker in the case of cell cultures. Our analysis shows that combining local cellular features with the properties of the cell's neighbourhood significantly improves the accuracy of machine learning-based phenotyping.

Original languageEnglish
Article number10085
JournalScientific Reports
Volume8
Issue number1
DOIs
Publication statusPublished - Dec 1 2018

Fingerprint

Cell culture
Learning systems
Cytology
Tissue
Image analysis
Microscopes
Imaging techniques
Chemical analysis

ASJC Scopus subject areas

  • General

Cite this

Toth, T., Balassa, T., Bara, N., Kovacs, F., Kriston, A., Molnar, C., ... Horvath, P. (2018). Environmental properties of cells improve machine learning-based phenotype recognition accuracy. Scientific Reports, 8(1), [10085]. https://doi.org/10.1038/s41598-018-28482-y

Environmental properties of cells improve machine learning-based phenotype recognition accuracy. / Toth, Timea; Balassa, Tamas; Bara, Norbert; Kovacs, Ferenc; Kriston, Andras; Molnar, Csaba; Haracska, L.; Sukosd, Farkas; Horvath, Peter.

In: Scientific Reports, Vol. 8, No. 1, 10085, 01.12.2018.

Research output: Contribution to journalArticle

Toth, T, Balassa, T, Bara, N, Kovacs, F, Kriston, A, Molnar, C, Haracska, L, Sukosd, F & Horvath, P 2018, 'Environmental properties of cells improve machine learning-based phenotype recognition accuracy', Scientific Reports, vol. 8, no. 1, 10085. https://doi.org/10.1038/s41598-018-28482-y
Toth, Timea ; Balassa, Tamas ; Bara, Norbert ; Kovacs, Ferenc ; Kriston, Andras ; Molnar, Csaba ; Haracska, L. ; Sukosd, Farkas ; Horvath, Peter. / Environmental properties of cells improve machine learning-based phenotype recognition accuracy. In: Scientific Reports. 2018 ; Vol. 8, No. 1.
@article{b5da5b690dce481c9e97a7673412a702,
title = "Environmental properties of cells improve machine learning-based phenotype recognition accuracy",
abstract = "To answer major questions of cell biology, it is often essential to understand the complex phenotypic composition of cellular systems precisely. Modern automated microscopes produce vast amounts of images routinely, making manual analysis nearly impossible. Due to their efficiency, machine learning-based analysis software have become essential tools to perform single-cell-level phenotypic analysis of large imaging datasets. However, an important limitation of such methods is that they do not use the information gained from the cellular micro- and macroenvironment: the algorithmic decision is based solely on the local properties of the cell of interest. Here, we present how various features from the surrounding environment contribute to identifying a cell and how such additional information can improve single-cell-level phenotypic image analysis. The proposed methodology was tested for different sizes of Euclidean and nearest neighbour-based cellular environments both on tissue sections and cell cultures. Our experimental data verify that the surrounding area of a cell largely determines its entity. This effect was found to be especially strong for established tissues, while it was somewhat weaker in the case of cell cultures. Our analysis shows that combining local cellular features with the properties of the cell's neighbourhood significantly improves the accuracy of machine learning-based phenotyping.",
author = "Timea Toth and Tamas Balassa and Norbert Bara and Ferenc Kovacs and Andras Kriston and Csaba Molnar and L. Haracska and Farkas Sukosd and Peter Horvath",
year = "2018",
month = "12",
day = "1",
doi = "10.1038/s41598-018-28482-y",
language = "English",
volume = "8",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",
number = "1",

}

TY - JOUR

T1 - Environmental properties of cells improve machine learning-based phenotype recognition accuracy

AU - Toth, Timea

AU - Balassa, Tamas

AU - Bara, Norbert

AU - Kovacs, Ferenc

AU - Kriston, Andras

AU - Molnar, Csaba

AU - Haracska, L.

AU - Sukosd, Farkas

AU - Horvath, Peter

PY - 2018/12/1

Y1 - 2018/12/1

N2 - To answer major questions of cell biology, it is often essential to understand the complex phenotypic composition of cellular systems precisely. Modern automated microscopes produce vast amounts of images routinely, making manual analysis nearly impossible. Due to their efficiency, machine learning-based analysis software have become essential tools to perform single-cell-level phenotypic analysis of large imaging datasets. However, an important limitation of such methods is that they do not use the information gained from the cellular micro- and macroenvironment: the algorithmic decision is based solely on the local properties of the cell of interest. Here, we present how various features from the surrounding environment contribute to identifying a cell and how such additional information can improve single-cell-level phenotypic image analysis. The proposed methodology was tested for different sizes of Euclidean and nearest neighbour-based cellular environments both on tissue sections and cell cultures. Our experimental data verify that the surrounding area of a cell largely determines its entity. This effect was found to be especially strong for established tissues, while it was somewhat weaker in the case of cell cultures. Our analysis shows that combining local cellular features with the properties of the cell's neighbourhood significantly improves the accuracy of machine learning-based phenotyping.

AB - To answer major questions of cell biology, it is often essential to understand the complex phenotypic composition of cellular systems precisely. Modern automated microscopes produce vast amounts of images routinely, making manual analysis nearly impossible. Due to their efficiency, machine learning-based analysis software have become essential tools to perform single-cell-level phenotypic analysis of large imaging datasets. However, an important limitation of such methods is that they do not use the information gained from the cellular micro- and macroenvironment: the algorithmic decision is based solely on the local properties of the cell of interest. Here, we present how various features from the surrounding environment contribute to identifying a cell and how such additional information can improve single-cell-level phenotypic image analysis. The proposed methodology was tested for different sizes of Euclidean and nearest neighbour-based cellular environments both on tissue sections and cell cultures. Our experimental data verify that the surrounding area of a cell largely determines its entity. This effect was found to be especially strong for established tissues, while it was somewhat weaker in the case of cell cultures. Our analysis shows that combining local cellular features with the properties of the cell's neighbourhood significantly improves the accuracy of machine learning-based phenotyping.

UR - http://www.scopus.com/inward/record.url?scp=85049651732&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85049651732&partnerID=8YFLogxK

U2 - 10.1038/s41598-018-28482-y

DO - 10.1038/s41598-018-28482-y

M3 - Article

AN - SCOPUS:85049651732

VL - 8

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

IS - 1

M1 - 10085

ER -