Uncovering the hidden half of plants using new advances in root phenotyping

Atkinson, Jonathan A.; Pound, Michael P.; Bennett, Malcolm; Wells, Darren M.

doi:10.1016/j.copbio.2018.06.002

Cited by 276 publications

(255 citation statements)

References 66 publications

Supporting

Mentioning

241

Contrasting

Order By: Relevance

“…Dynamics are challenging to phenotype because space and time dimensions are required. Phenotyping in 4D can be done with magnetic resonance imaging (MRI), X-ray computed tomography (CT), and positron emission tomography (PET) [38]. These technologies can measure root growth in different soil types and in undisturbed soil cores [39][40][41], and in response to phosphorus [42] and water [43].…”

Section: Functional and Dynamic Phenotypes For Capture Of Soil Resourcesmentioning

confidence: 99%

Crop Improvement from Phenotyping Roots: Highlights Reveal Expanding Opportunities

Tracy

Nagel

Postma

et al. 2020

Trends in Plant Science

182

128

View full text Add to dashboard Cite

Root systems determine the water and nutrients for photosynthesis and harvested products, underpinning agricultural productivity. We highlight 11 programs that integrated root traits into germplasm for breeding, relying on phenotyping. Progress was successful but slow. Today's phenotyping technologies will speed up root trait improvement. They combine multiple new alleles in germplasm for target environments, in parallel. Roots and shoots are detected simultaneously and nondestructively, seed to seed measures are automated, and field and laboratory technologies are increasingly linked. Available simulation models can aid all phenotyping decisions. This century will see a shift from single root traits to rhizosphere selections that can be managed dynamically on farms and a shift to phenotype-based improvement to accommodate the dynamic complexity of whole crop systems.Root system traits (see Glossary) have long been a key target by researchers and breeders for crop improvement [1,2]. Root system architecture supplies water and nutrients for photosynthesis and growth, stabilizes the plant, and prevents soil toxic elements and pathogens from entering leaves and reproductive organs. Roots also host soil microorganisms that can contribute to plant growth and resource efficiency and modulate the supply of resources and signals to shoots, influencing partitioning to organs, including flowers, seed, and fruit. Despite the challenges that plant roots present for measurement, root traits have been incorporated into crops using phenotyping. The observation of roots using phenotyping is central to the discovery of root traits beneficial to crops, their incorporation into new cultivars using prebreeding [3,4], and to their management using precision agriculture. Highlights in Root PhenotypingThe 11 programs highlighted in Figure 1 (Key Figure) exemplify root traits incorporated into new genotypes by phenotyping to increase crop productivity. The examples cover the two main plant types and root system developments: monocotyledons (two major cereal crops, wheat and rice) with seed and stem-borne roots with no cambial thickening, and dicotyledons (two major legume crops, bean

show abstract

Section: Functional and Dynamic Phenotypes For Capture Of Soil Resourcesmentioning

confidence: 99%

Crop Improvement from Phenotyping Roots: Highlights Reveal Expanding Opportunities

Tracy

Nagel

Postma

et al. 2020

Trends in Plant Science

182

128

View full text Add to dashboard Cite

show abstract

“…In this paper, we focus on the analysis of root systems where improvements promise increases to water and nutrient use e ciency [8]. Historically, automated root phenotyping has proven challenging, due partly to the concealed nature of roots in the soil, but also to the architectural complexity and variability of root systems between species, and even individuals.…”

Section: Analysis Of Root System Architecturesmentioning

confidence: 99%

RootNav 2.0: Deep Learning for Automatic Navigation of Complex Plant Root Architectures

Yasrab

Atkinson

Wells

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

We present a new image analysis approach that provides fully-automatic extraction of complex root system architectures from a range of plant species in varied imaging setups. Driven by modern deep-learning approaches, RootNav 2.0 replaces previously manual and semi-automatic feature extraction with an extremely deep multi-task Convolutional Neural Network architecture. The network has been designed to explicitly combine local pixel information with global scene information in order to accurately segment small root features across high-resolution images. In addition, the network simultaneously locates seeds, and rst and second order root tips to drive a search algorithm seeking optimal paths throughout the image, extracting accurate architectures without user interaction. The proposed method is evaluated on images of wheat (Triticum aestivum L.) from a seedling assay. The results are compared with semi-automatic analysis via the original RootNav tool, demonstrating comparable accuracy, with a 10-fold increase in speed. We then demonstrate the ability of the network to adapt to di erent plant species via transfer learning, o ering similar accuracy when transferred to an Arabidopsis thaliana plate assay. We transfer for a nal time to images of Brassica napus from a hydroponic assay, and still demonstrate good accuracy despite many fewer training images. The tool outputs root architectures in the widely accepted RSML standard, for which numerous analysis packages exist (http://rootsystemml.github.io/), as well as segmentation masks compatible with other automated measurement tools.

show abstract

“…Recent developments in root-soil imaging has enabled the direct examination of dynamic plant-soil interactions [4], but the data generated poses significant technological challenges. X-ray Computed Tomography (CT) is one leading technology for obtaining non-destructive root images without disturbing the root or soil structure [5]. However, a major bottleneck in the study of root systems using CT is the computational analysis of the large volumetric images produced; analysing CT data is a very time-consuming task.…”

Section: Introductionmentioning

confidence: 99%

Three Dimensional Root CT Segmentation using Multi-Resolution Encoder-Decoder Networks

Soltaninejad

Sturrock

Griffiths

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

We address the complex problem of reliably segmenting root structure from soil in X-ray Computed Tomography (CT) images. We utilise a deep learning approach, and propose a state-of-the-art multi-resolution architecture based on encoder-decoders. While previous work in encoder-decoders implies the use of multiple resolutions simply by downsampling and upsampling images, we make this process explicit, with branches of the network tasked separately with obtaining local high-resolution segmentation, and wider low-resolution contextual information. The complete network is a memory efficient implementation that is still able to resolve small root detail in large volumetric images. We evaluate our approach by comparing against a number of different encoder-decoder based architectures from the literature, as well as a popular existing image analysis tool designed for root CT segmentation. We show qualitatively and quantitatively that a multi-resolution approach offers substantial accuracy improvements over a both a small receptive field size in a deep network, or a larger receptive field in a shallower network. We obtain a Dice score of 0.59 compared with 0.41 for the closest competing method. We then further improve performance using an incremental learning approach, in which failures in the original network are used to generate harder negative training examples. Results of this process raise the precision of the network, and improve the Dice score to 0.66. Our proposed method requires no user interaction, is fully automatic, and identifies large and fine root material throughout the whole volume. The 3D segmented output of our method is well-connected, allowing the recovery of structured representations of root system architecture, and so may be successfully utilised in root phenotyping.

show abstract

Uncovering the hidden half of plants using new advances in root phenotyping

Cited by 276 publications

References 66 publications

Crop Improvement from Phenotyping Roots: Highlights Reveal Expanding Opportunities

Crop Improvement from Phenotyping Roots: Highlights Reveal Expanding Opportunities

RootNav 2.0: Deep Learning for Automatic Navigation of Complex Plant Root Architectures

Three Dimensional Root CT Segmentation using Multi-Resolution Encoder-Decoder Networks

Contact Info

Product

Resources

About