Converting tabular data into images for deep learning with convolutional neural networks

Abstract Convolutional neural networks (CNNs) have been successfully used in many applications where important information about data is embedded in the order of features, such as speech and imaging. However, most tabular data do not assume a spatial relationship between features, and thus are unsui...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Yitan Zhu, Thomas Brettin, Fangfang Xia, Alexander Partin, Maulik Shukla, Hyunseung Yoo, Yvonne A. Evrard, James H. Doroshow, Rick L. Stevens
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/37c2f916e6fa45c08315225e5a9e05de
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:37c2f916e6fa45c08315225e5a9e05de
record_format	dspace
spelling	oai:doaj.org-article:37c2f916e6fa45c08315225e5a9e05de2021-12-02T18:24:53ZConverting tabular data into images for deep learning with convolutional neural networks10.1038/s41598-021-90923-y2045-2322https://doaj.org/article/37c2f916e6fa45c08315225e5a9e05de2021-05-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-90923-yhttps://doaj.org/toc/2045-2322Abstract Convolutional neural networks (CNNs) have been successfully used in many applications where important information about data is embedded in the order of features, such as speech and imaging. However, most tabular data do not assume a spatial relationship between features, and thus are unsuitable for modeling using CNNs. To meet this challenge, we develop a novel algorithm, image generator for tabular data (IGTD), to transform tabular data into images by assigning features to pixel positions so that similar features are close to each other in the image. The algorithm searches for an optimized assignment by minimizing the difference between the ranking of distances between features and the ranking of distances between their assigned pixels in the image. We apply IGTD to transform gene expression profiles of cancer cell lines (CCLs) and molecular descriptors of drugs into their respective image representations. Compared with existing transformation methods, IGTD generates compact image representations with better preservation of feature neighborhood structure. Evaluated on benchmark drug screening datasets, CNNs trained on IGTD image representations of CCLs and drugs exhibit a better performance of predicting anti-cancer drug response than both CNNs trained on alternative image representations and prediction models trained on the original tabular data.Yitan ZhuThomas BrettinFangfang XiaAlexander PartinMaulik ShuklaHyunseung YooYvonne A. EvrardJames H. DoroshowRick L. StevensNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Yitan Zhu Thomas Brettin Fangfang Xia Alexander Partin Maulik Shukla Hyunseung Yoo Yvonne A. Evrard James H. Doroshow Rick L. Stevens Converting tabular data into images for deep learning with convolutional neural networks
description	Abstract Convolutional neural networks (CNNs) have been successfully used in many applications where important information about data is embedded in the order of features, such as speech and imaging. However, most tabular data do not assume a spatial relationship between features, and thus are unsuitable for modeling using CNNs. To meet this challenge, we develop a novel algorithm, image generator for tabular data (IGTD), to transform tabular data into images by assigning features to pixel positions so that similar features are close to each other in the image. The algorithm searches for an optimized assignment by minimizing the difference between the ranking of distances between features and the ranking of distances between their assigned pixels in the image. We apply IGTD to transform gene expression profiles of cancer cell lines (CCLs) and molecular descriptors of drugs into their respective image representations. Compared with existing transformation methods, IGTD generates compact image representations with better preservation of feature neighborhood structure. Evaluated on benchmark drug screening datasets, CNNs trained on IGTD image representations of CCLs and drugs exhibit a better performance of predicting anti-cancer drug response than both CNNs trained on alternative image representations and prediction models trained on the original tabular data.
format	article
author	Yitan Zhu Thomas Brettin Fangfang Xia Alexander Partin Maulik Shukla Hyunseung Yoo Yvonne A. Evrard James H. Doroshow Rick L. Stevens
author_facet	Yitan Zhu Thomas Brettin Fangfang Xia Alexander Partin Maulik Shukla Hyunseung Yoo Yvonne A. Evrard James H. Doroshow Rick L. Stevens
author_sort	Yitan Zhu
title	Converting tabular data into images for deep learning with convolutional neural networks
title_short	Converting tabular data into images for deep learning with convolutional neural networks
title_full	Converting tabular data into images for deep learning with convolutional neural networks
title_fullStr	Converting tabular data into images for deep learning with convolutional neural networks
title_full_unstemmed	Converting tabular data into images for deep learning with convolutional neural networks
title_sort	converting tabular data into images for deep learning with convolutional neural networks
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/37c2f916e6fa45c08315225e5a9e05de
work_keys_str_mv	AT yitanzhu convertingtabulardataintoimagesfordeeplearningwithconvolutionalneuralnetworks AT thomasbrettin convertingtabulardataintoimagesfordeeplearningwithconvolutionalneuralnetworks AT fangfangxia convertingtabulardataintoimagesfordeeplearningwithconvolutionalneuralnetworks AT alexanderpartin convertingtabulardataintoimagesfordeeplearningwithconvolutionalneuralnetworks AT maulikshukla convertingtabulardataintoimagesfordeeplearningwithconvolutionalneuralnetworks AT hyunseungyoo convertingtabulardataintoimagesfordeeplearningwithconvolutionalneuralnetworks AT yvonneaevrard convertingtabulardataintoimagesfordeeplearningwithconvolutionalneuralnetworks AT jameshdoroshow convertingtabulardataintoimagesfordeeplearningwithconvolutionalneuralnetworks AT ricklstevens convertingtabulardataintoimagesfordeeplearningwithconvolutionalneuralnetworks
_version_	1718378088049410048

Converting tabular data into images for deep learning with convolutional neural networks

Ejemplares similares