Cargando…

Automated Counting of Cancer Cells by Ensembling Deep Features

High-content and high-throughput digital microscopes have generated large image sets in biological experiments and clinical practice. Automatic image analysis techniques, such as cell counting, are in high demand. Here, cell counting was treated as a regression problem using image features (phenotyp...

Descripción completa

Detalles Bibliográficos
Autores principales: Liu, Qian, Junker, Anna, Murakami, Kazuhiro, Hu, Pingzhao
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6770845/
https://www.ncbi.nlm.nih.gov/pubmed/31480740
http://dx.doi.org/10.3390/cells8091019
_version_ 1783455577848938496
author Liu, Qian
Junker, Anna
Murakami, Kazuhiro
Hu, Pingzhao
author_facet Liu, Qian
Junker, Anna
Murakami, Kazuhiro
Hu, Pingzhao
author_sort Liu, Qian
collection PubMed
description High-content and high-throughput digital microscopes have generated large image sets in biological experiments and clinical practice. Automatic image analysis techniques, such as cell counting, are in high demand. Here, cell counting was treated as a regression problem using image features (phenotypes) extracted by deep learning models. Three deep convolutional neural network models were developed to regress image features to their cell counts in an end-to-end way. Theoretically, ensembling imaging phenotypes should have better representative ability than a single type of imaging phenotype. We implemented this idea by integrating two types of imaging phenotypes (dot density map and foreground mask) extracted by two autoencoders and regressing the ensembled imaging phenotypes to cell counts afterwards. Two publicly available datasets with synthetic microscopic images were used to train and test the proposed models. Root mean square error, mean absolute error, mean absolute percent error, and Pearson correlation were applied to evaluate the models’ performance. The well-trained models were also applied to predict the cancer cell counts of real microscopic images acquired in a biological experiment to evaluate the roles of two colorectal-cancer-related genes. The proposed model by ensembling deep imaging features showed better performance in terms of smaller errors and larger correlations than those based on a single type of imaging feature. Overall, all models’ predictions showed a high correlation with the true cell counts. The ensembling-based model integrated high-level imaging phenotypes to improve the estimation of cell counts from high-content and high-throughput microscopic images.
format Online
Article
Text
id pubmed-6770845
institution National Center for Biotechnology Information
language English
publishDate 2019
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-67708452019-10-30 Automated Counting of Cancer Cells by Ensembling Deep Features Liu, Qian Junker, Anna Murakami, Kazuhiro Hu, Pingzhao Cells Article High-content and high-throughput digital microscopes have generated large image sets in biological experiments and clinical practice. Automatic image analysis techniques, such as cell counting, are in high demand. Here, cell counting was treated as a regression problem using image features (phenotypes) extracted by deep learning models. Three deep convolutional neural network models were developed to regress image features to their cell counts in an end-to-end way. Theoretically, ensembling imaging phenotypes should have better representative ability than a single type of imaging phenotype. We implemented this idea by integrating two types of imaging phenotypes (dot density map and foreground mask) extracted by two autoencoders and regressing the ensembled imaging phenotypes to cell counts afterwards. Two publicly available datasets with synthetic microscopic images were used to train and test the proposed models. Root mean square error, mean absolute error, mean absolute percent error, and Pearson correlation were applied to evaluate the models’ performance. The well-trained models were also applied to predict the cancer cell counts of real microscopic images acquired in a biological experiment to evaluate the roles of two colorectal-cancer-related genes. The proposed model by ensembling deep imaging features showed better performance in terms of smaller errors and larger correlations than those based on a single type of imaging feature. Overall, all models’ predictions showed a high correlation with the true cell counts. The ensembling-based model integrated high-level imaging phenotypes to improve the estimation of cell counts from high-content and high-throughput microscopic images. MDPI 2019-09-02 /pmc/articles/PMC6770845/ /pubmed/31480740 http://dx.doi.org/10.3390/cells8091019 Text en © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Liu, Qian
Junker, Anna
Murakami, Kazuhiro
Hu, Pingzhao
Automated Counting of Cancer Cells by Ensembling Deep Features
title Automated Counting of Cancer Cells by Ensembling Deep Features
title_full Automated Counting of Cancer Cells by Ensembling Deep Features
title_fullStr Automated Counting of Cancer Cells by Ensembling Deep Features
title_full_unstemmed Automated Counting of Cancer Cells by Ensembling Deep Features
title_short Automated Counting of Cancer Cells by Ensembling Deep Features
title_sort automated counting of cancer cells by ensembling deep features
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6770845/
https://www.ncbi.nlm.nih.gov/pubmed/31480740
http://dx.doi.org/10.3390/cells8091019
work_keys_str_mv AT liuqian automatedcountingofcancercellsbyensemblingdeepfeatures
AT junkeranna automatedcountingofcancercellsbyensemblingdeepfeatures
AT murakamikazuhiro automatedcountingofcancercellsbyensemblingdeepfeatures
AT hupingzhao automatedcountingofcancercellsbyensemblingdeepfeatures