Time-Series Growth Prediction Model Based on U-Net and Machine Learning in Arabidopsis

Yield prediction for crops is essential information for food security. A high-throughput phenotyping platform (HTPP) generates the data of the complete life cycle of a plant. However, the data are rarely used for yield prediction because of the lack of quality image analysis methods, yield data asso...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Sungyul Chang, Unseok Lee, Min Jeong Hong, Yeong Deuk Jo, Jin-Baek Kim
Formato:	article
Lenguaje:	EN
Publicado:	Frontiers Media S.A. 2021
Materias:	time series analysis phenomics high-throughput phenotyping (HTP) deep learning DL) growth modeling plant biomass Plant culture SB1-1110
Acceso en línea:	https://doaj.org/article/100486e6f4ed4036bd348f4e7ee2be2c
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:100486e6f4ed4036bd348f4e7ee2be2c
record_format	dspace
spelling	oai:doaj.org-article:100486e6f4ed4036bd348f4e7ee2be2c2021-11-11T05:33:15ZTime-Series Growth Prediction Model Based on U-Net and Machine Learning in Arabidopsis1664-462X10.3389/fpls.2021.721512https://doaj.org/article/100486e6f4ed4036bd348f4e7ee2be2c2021-11-01T00:00:00Zhttps://www.frontiersin.org/articles/10.3389/fpls.2021.721512/fullhttps://doaj.org/toc/1664-462XYield prediction for crops is essential information for food security. A high-throughput phenotyping platform (HTPP) generates the data of the complete life cycle of a plant. However, the data are rarely used for yield prediction because of the lack of quality image analysis methods, yield data associated with HTPP, and the time-series analysis method for yield prediction. To overcome limitations, this study employed multiple deep learning (DL) networks to extract high-quality HTTP data, establish an association between HTTP data and the yield performance of crops, and select essential time intervals using machine learning (ML). The images of Arabidopsis were taken 12 times under environmentally controlled HTPP over 23 days after sowing (DAS). First, the features from images were extracted using DL network U-Net with SE-ResXt101 encoder and divided into early (15–21 DAS) and late (∼21–23 DAS) pre-flowering developmental stages using the physiological characteristics of the Arabidopsis plant. Second, the late pre-flowering stage at 23 DAS can be predicted using the ML algorithm XGBoost, based only on a portion of the early pre-flowering stage (17–21 DAS). This was confirmed using an additional biological experiment (P < 0.01). Finally, the projected area (PA) was estimated into fresh weight (FW), and the correlation coefficient between FW and predicted FW was calculated as 0.85. This was the first study that analyzed time-series data to predict the FW of related but different developmental stages and predict the PA. The results of this study were informative and enabled the understanding of the FW of Arabidopsis or yield of leafy plants and total biomass consumed in vertical farming. Moreover, this study highlighted the reduction of time-series data for examining interesting traits and future application of time-series analysis in various HTPPs.Sungyul ChangUnseok LeeMin Jeong HongYeong Deuk JoJin-Baek KimFrontiers Media S.A.articletime series analysisphenomicshigh-throughput phenotyping (HTP)deep learning DL)growth modelingplant biomassPlant cultureSB1-1110ENFrontiers in Plant Science, Vol 12 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	time series analysis phenomics high-throughput phenotyping (HTP) deep learning DL) growth modeling plant biomass Plant culture SB1-1110
spellingShingle	time series analysis phenomics high-throughput phenotyping (HTP) deep learning DL) growth modeling plant biomass Plant culture SB1-1110 Sungyul Chang Unseok Lee Min Jeong Hong Yeong Deuk Jo Jin-Baek Kim Time-Series Growth Prediction Model Based on U-Net and Machine Learning in Arabidopsis
description	Yield prediction for crops is essential information for food security. A high-throughput phenotyping platform (HTPP) generates the data of the complete life cycle of a plant. However, the data are rarely used for yield prediction because of the lack of quality image analysis methods, yield data associated with HTPP, and the time-series analysis method for yield prediction. To overcome limitations, this study employed multiple deep learning (DL) networks to extract high-quality HTTP data, establish an association between HTTP data and the yield performance of crops, and select essential time intervals using machine learning (ML). The images of Arabidopsis were taken 12 times under environmentally controlled HTPP over 23 days after sowing (DAS). First, the features from images were extracted using DL network U-Net with SE-ResXt101 encoder and divided into early (15–21 DAS) and late (∼21–23 DAS) pre-flowering developmental stages using the physiological characteristics of the Arabidopsis plant. Second, the late pre-flowering stage at 23 DAS can be predicted using the ML algorithm XGBoost, based only on a portion of the early pre-flowering stage (17–21 DAS). This was confirmed using an additional biological experiment (P < 0.01). Finally, the projected area (PA) was estimated into fresh weight (FW), and the correlation coefficient between FW and predicted FW was calculated as 0.85. This was the first study that analyzed time-series data to predict the FW of related but different developmental stages and predict the PA. The results of this study were informative and enabled the understanding of the FW of Arabidopsis or yield of leafy plants and total biomass consumed in vertical farming. Moreover, this study highlighted the reduction of time-series data for examining interesting traits and future application of time-series analysis in various HTPPs.
format	article
author	Sungyul Chang Unseok Lee Min Jeong Hong Yeong Deuk Jo Jin-Baek Kim
author_facet	Sungyul Chang Unseok Lee Min Jeong Hong Yeong Deuk Jo Jin-Baek Kim
author_sort	Sungyul Chang
title	Time-Series Growth Prediction Model Based on U-Net and Machine Learning in Arabidopsis
title_short	Time-Series Growth Prediction Model Based on U-Net and Machine Learning in Arabidopsis
title_full	Time-Series Growth Prediction Model Based on U-Net and Machine Learning in Arabidopsis
title_fullStr	Time-Series Growth Prediction Model Based on U-Net and Machine Learning in Arabidopsis
title_full_unstemmed	Time-Series Growth Prediction Model Based on U-Net and Machine Learning in Arabidopsis
title_sort	time-series growth prediction model based on u-net and machine learning in arabidopsis
publisher	Frontiers Media S.A.
publishDate	2021
url	https://doaj.org/article/100486e6f4ed4036bd348f4e7ee2be2c
work_keys_str_mv	AT sungyulchang timeseriesgrowthpredictionmodelbasedonunetandmachinelearninginarabidopsis AT unseoklee timeseriesgrowthpredictionmodelbasedonunetandmachinelearninginarabidopsis AT minjeonghong timeseriesgrowthpredictionmodelbasedonunetandmachinelearninginarabidopsis AT yeongdeukjo timeseriesgrowthpredictionmodelbasedonunetandmachinelearninginarabidopsis AT jinbaekkim timeseriesgrowthpredictionmodelbasedonunetandmachinelearninginarabidopsis
_version_	1718439465383362560

Time-Series Growth Prediction Model Based on U-Net and Machine Learning in Arabidopsis

Ejemplares similares