Dataset Information

Doing More With Less: A Multitask Deep Learning Approach in Plant Phenotyping.

ABSTRACT: Image-based plant phenotyping has been steadily growing and this has steeply increased the need for more efficient image analysis techniques capable of evaluating multiple plant traits. Deep learning has shown its potential in a multitude of visual tasks in plant phenotyping, such as segmentation and counting. Here, we show how different phenotyping traits can be extracted simultaneously from plant images, using multitask learning (MTL). MTL leverages information contained in the training images of related tasks to improve overall generalization and learns models with fewer labels. We present a multitask deep learning framework for plant phenotyping, able to infer three traits simultaneously: (i) leaf count, (ii) projected leaf area (PLA), and (iii) genotype classification. We adopted a modified pretrained ResNet50 as a feature extractor, trained end-to-end to predict multiple traits. We also leverage MTL to show that through learning from more easily obtainable annotations (such as PLA and genotype) we can predict a better leaf count (harder to obtain annotation). We evaluate our findings on several publicly available datasets of top-view images of Arabidopsis thaliana. Experimental results show that the proposed MTL method improves the leaf count mean squared error (MSE) by more than 40%, compared to a single task network on the same dataset. We also show that our MTL framework can be trained with up to 75% fewer leaf count annotations without significantly impacting performance, whereas a single task model shows a steady decline when fewer annotations are available. Code available at https://github.com/andobrescu/Multi_task_plant_phenotyping.

SUBMITTER: Dobrescu A

PROVIDER: S-EPMC7093010 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Doing More With Less: A Multitask Deep Learning Approach in Plant Phenotyping.

Dobrescu Andrei A Giuffrida Mario Valerio MV Tsaftaris Sotirios A SA

Frontiers in plant science 20200228

Image-based plant phenotyping has been steadily growing and this has steeply increased the need for more efficient image analysis techniques capable of evaluating multiple plant traits. Deep learning has shown its potential in a multitude of visual tasks in plant phenotyping, such as segmentation and counting. Here, we show how different phenotyping traits can be extracted simultaneously from plant images, using multitask learning (MTL). MTL leverages information contained in the training images ...[more]

PMID: 32256503

Similar Datasets

Project description:ObjectivesBubble CPAP (bCPAP), a non-invasive ventilation modality, has emerged as an intervention that is able to reduce pneumonia-related mortality in children in low resourced settings. Our study primarily aimed to describe a cohort of children who were started on CPAP in the Medical Emergency Unit (MEU) of Red Cross War Memorial Children's Hospital 2016-2018.MethodsA retrospective review of a randomly selected sample of paper-based folders was conducted. Children started on bCPAP at MEU were eligible for inclusion. Demographic and clinical data, management, and outcomes regarding admission to PICU, need for invasive ventilation and mortality were documented. Descriptive statistical data were generated for all relevant variables. Percentages depicted frequencies of categorical data while medians with interquartile ranges (IQR) were used to summarise continuous data.ResultsOf 500 children started on bCPAP, 266 (53%) were male; their median age was 3.7 (IQR 1.7-11.3) months and 169 (34%) were moderately to severely underweight-for-age. There were 12 (2%) HIV-infected children; 403 (81%) had received appropriate immunisations for their age; and 119 (24%) were exposed to tobacco smoke at home. The five most common primary reasons for admission were acute respiratory illness, acute gastroenteritis, congestive cardiac failure, sepsis and seizures. Most children, 409 (82%), had no underlying medical condition. Most children, 411 (82%), were managed in high care areas of the general medical wards while 126 (25%) went to PICU. The median time on CPAP was 1.7 (IQR 0.9-2.8) days. The median hospitalisation time was 6 (IQR 4-9) days. Overall, 38 (8%) children required invasive ventilatory support. Overall, 12 (2%) children with a median age of 7.5 (IQR 0.7-14.5) months died, six of whom had an underlying medical condition.ConclusionsSeventy-five percent of children initiated on bCPAP did not require PICU admission. This form of non-invasive ventilatory support should be considered more widely in the context of limited access to paediatric intensive care units in other African settings.

Project description:BackgroundMaize (Zea mays L.) is one of the most important food sources in the world and has been one of the main targets of plant genetics and phenotypic research for centuries. Observation and analysis of various morphological phenotypic traits during maize growth are essential for genetic and breeding study. The generally huge number of samples produce an enormous amount of high-resolution image data. While high throughput plant phenotyping platforms are increasingly used in maize breeding trials, there is a reasonable need for software tools that can automatically identify visual phenotypic features of maize plants and implement batch processing on image datasets.ResultsOn the boundary between computer vision and plant science, we utilize advanced deep learning methods based on convolutional neural networks to empower the workflow of maize phenotyping analysis. This paper presents Maize-IAS (Maize Image Analysis Software), an integrated application supporting one-click analysis of maize phenotype, embedding multiple functions: (I) Projection, (II) Color Analysis, (III) Internode length, (IV) Height, (V) Stem Diameter and (VI) Leaves Counting. Taking the RGB image of maize as input, the software provides a user-friendly graphical interaction interface and rapid calculation of multiple important phenotypic characteristics, including leaf sheath points detection and leaves segmentation. In function Leaves Counting, the mean and standard deviation of difference between prediction and ground truth are 1.60 and 1.625.ConclusionThe Maize-IAS is easy-to-use and demands neither professional knowledge of computer vision nor deep learning. All functions for batch processing are incorporated, enabling automated and labor-reduced tasks of recording, measurement and quantitative analysis of maize growth traits on a large dataset. We prove the efficiency and potential capability of our techniques and software to image-based plant research, which also demonstrates the feasibility and capability of AI technology implemented in agriculture and plant science.

Dataset Information

Doing More With Less: A Multitask Deep Learning Approach in Plant Phenotyping.

Publications

Doing More With Less: A Multitask Deep Learning Approach in Plant Phenotyping.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets