Label-free optical hemogram of granulocytes enhanced by artificial neural networks

Roopam K. Gupta; Mingzhou Chen; Graeme P.A. Malcolm; Nils Hempler; Kishan Dholakia; Simon J. Powis

doi:10.1364/OE.27.013706

Label-free optical hemogram of granulocytes enhanced by artificial neural networks

Roopam K. Gupta, Mingzhou Chen, Graeme P.A. Malcolm, Nils Hempler, Kishan Dholakia, Simon J. Powis

Department of Physics

Research output: Contribution to journal › Article › peer-review

7 Citations (Scopus)

Abstract

An outstanding challenge for immunology is the classification of immune cells in a label-free fashion with high speed. For this purpose, optical techniques such as Raman spectroscopy or digital holographic microscopy have been used successfully to identify immune cell subsets. To achieve high accuracy, these techniques require a post-processing step using linear methods of multivariate processing, such as principal component analysis. Here we demonstrate for the first time a comparison between artificial neural networks and principal component analysis (PCA) to classify the key granulocyte cell lineages of neutrophils and eosinophils using both digital holographic microscopy and Raman spectroscopy. Artificial neural networks can offer advantages in terms of classification accuracy and speed over a PCA approach. We conclude that digital holographic microscopy with convolutional neural networks based analysis provides a route to a robust, stand-alone and high-throughput hemogram with a classification accuracy of 91.3 % at a throughput rate of greater than 100 cells per second.

Original language	English
Pages (from-to)	13706-13720
Number of pages	15
Journal	Optics Express
Volume	27
Issue number	10
DOIs	https://doi.org/10.1364/OE.27.013706
Publication status	Published - 2019

Bibliographical note

Funding Information:
This work was supported by a Medical Research Scotland PhD studentship awarded to R.K.G, and grant funding from the UK Engineering and Physical Sciences Research Council (grants EP/R004854/1, EP/P030017/1). The opinions expressed in this article are the authors own and do not reflect the view of above mentioned funding agencies. KD and SJP developed the project. RKG developed the numerical analysis procedures presented and performed the experiments with MC and SJP. RKG, MC, SJP and KD wrote the paper which was approved by NH and GPAM. KD, SJP and NH supervised the project. R. Gupta, M. Chen, G. P. A. Malcolm, N. Hempler, K. Dholakia, and S. J. Powis provide the following: Data Underpinning: A label-free optical hemogram of granulocytes enhanced by artificial neural networks, Dataset, University of St Andrews Research Portal, https://doi.org/10.17630/c3b0856b-3400-4211-9e7c-eacfc7082067.

Publisher Copyright:
© 2019 OSA - The Optical Society. All rights reserved.

All Science Journal Classification (ASJC) codes

Atomic and Molecular Physics, and Optics

Access to Document

10.1364/OE.27.013706

Cite this

@article{7b18fd84f10b48d8ba30aa22fff86084,

title = "Label-free optical hemogram of granulocytes enhanced by artificial neural networks",

abstract = "An outstanding challenge for immunology is the classification of immune cells in a label-free fashion with high speed. For this purpose, optical techniques such as Raman spectroscopy or digital holographic microscopy have been used successfully to identify immune cell subsets. To achieve high accuracy, these techniques require a post-processing step using linear methods of multivariate processing, such as principal component analysis. Here we demonstrate for the first time a comparison between artificial neural networks and principal component analysis (PCA) to classify the key granulocyte cell lineages of neutrophils and eosinophils using both digital holographic microscopy and Raman spectroscopy. Artificial neural networks can offer advantages in terms of classification accuracy and speed over a PCA approach. We conclude that digital holographic microscopy with convolutional neural networks based analysis provides a route to a robust, stand-alone and high-throughput hemogram with a classification accuracy of 91.3 % at a throughput rate of greater than 100 cells per second.",

author = "Gupta, {Roopam K.} and Mingzhou Chen and Malcolm, {Graeme P.A.} and Nils Hempler and Kishan Dholakia and Powis, {Simon J.}",

note = "Funding Information: This work was supported by a Medical Research Scotland PhD studentship awarded to R.K.G, and grant funding from the UK Engineering and Physical Sciences Research Council (grants EP/R004854/1, EP/P030017/1). The opinions expressed in this article are the authors own and do not reflect the view of above mentioned funding agencies. KD and SJP developed the project. RKG developed the numerical analysis procedures presented and performed the experiments with MC and SJP. RKG, MC, SJP and KD wrote the paper which was approved by NH and GPAM. KD, SJP and NH supervised the project. R. Gupta, M. Chen, G. P. A. Malcolm, N. Hempler, K. Dholakia, and S. J. Powis provide the following: Data Underpinning: A label-free optical hemogram of granulocytes enhanced by artificial neural networks, Dataset, University of St Andrews Research Portal, https://doi.org/10.17630/c3b0856b-3400-4211-9e7c-eacfc7082067. Publisher Copyright: {\textcopyright} 2019 OSA - The Optical Society. All rights reserved.",

year = "2019",

doi = "10.1364/OE.27.013706",

language = "English",

volume = "27",

pages = "13706--13720",

journal = "Optics Express",

issn = "1094-4087",

publisher = "The Optical Society",

number = "10",

}

TY - JOUR

T1 - Label-free optical hemogram of granulocytes enhanced by artificial neural networks

AU - Gupta, Roopam K.

AU - Chen, Mingzhou

AU - Malcolm, Graeme P.A.

AU - Hempler, Nils

AU - Dholakia, Kishan

AU - Powis, Simon J.

N1 - Funding Information: This work was supported by a Medical Research Scotland PhD studentship awarded to R.K.G, and grant funding from the UK Engineering and Physical Sciences Research Council (grants EP/R004854/1, EP/P030017/1). The opinions expressed in this article are the authors own and do not reflect the view of above mentioned funding agencies. KD and SJP developed the project. RKG developed the numerical analysis procedures presented and performed the experiments with MC and SJP. RKG, MC, SJP and KD wrote the paper which was approved by NH and GPAM. KD, SJP and NH supervised the project. R. Gupta, M. Chen, G. P. A. Malcolm, N. Hempler, K. Dholakia, and S. J. Powis provide the following: Data Underpinning: A label-free optical hemogram of granulocytes enhanced by artificial neural networks, Dataset, University of St Andrews Research Portal, https://doi.org/10.17630/c3b0856b-3400-4211-9e7c-eacfc7082067. Publisher Copyright: © 2019 OSA - The Optical Society. All rights reserved.

PY - 2019

Y1 - 2019

N2 - An outstanding challenge for immunology is the classification of immune cells in a label-free fashion with high speed. For this purpose, optical techniques such as Raman spectroscopy or digital holographic microscopy have been used successfully to identify immune cell subsets. To achieve high accuracy, these techniques require a post-processing step using linear methods of multivariate processing, such as principal component analysis. Here we demonstrate for the first time a comparison between artificial neural networks and principal component analysis (PCA) to classify the key granulocyte cell lineages of neutrophils and eosinophils using both digital holographic microscopy and Raman spectroscopy. Artificial neural networks can offer advantages in terms of classification accuracy and speed over a PCA approach. We conclude that digital holographic microscopy with convolutional neural networks based analysis provides a route to a robust, stand-alone and high-throughput hemogram with a classification accuracy of 91.3 % at a throughput rate of greater than 100 cells per second.

AB - An outstanding challenge for immunology is the classification of immune cells in a label-free fashion with high speed. For this purpose, optical techniques such as Raman spectroscopy or digital holographic microscopy have been used successfully to identify immune cell subsets. To achieve high accuracy, these techniques require a post-processing step using linear methods of multivariate processing, such as principal component analysis. Here we demonstrate for the first time a comparison between artificial neural networks and principal component analysis (PCA) to classify the key granulocyte cell lineages of neutrophils and eosinophils using both digital holographic microscopy and Raman spectroscopy. Artificial neural networks can offer advantages in terms of classification accuracy and speed over a PCA approach. We conclude that digital holographic microscopy with convolutional neural networks based analysis provides a route to a robust, stand-alone and high-throughput hemogram with a classification accuracy of 91.3 % at a throughput rate of greater than 100 cells per second.

UR - http://www.scopus.com/inward/record.url?scp=85065831289&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065831289&partnerID=8YFLogxK

U2 - 10.1364/OE.27.013706

DO - 10.1364/OE.27.013706

M3 - Article

C2 - 31163830

AN - SCOPUS:85065831289

SN - 1094-4087

VL - 27

SP - 13706

EP - 13720

JO - Optics Express

JF - Optics Express

IS - 10

ER -

Label-free optical hemogram of granulocytes enhanced by artificial neural networks

Abstract

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this