RooTrak: Automated Recovery of Three-Dimensional Plant Root Architecture in Soil from X-Ray Microcomputed Tomography Images Using Visual Tracking

Centre for Plant Integrative Biology, University of Nottingham, Nottingham LE12 5RD, United Kingdom.
Plant physiology (Impact Factor: 6.84). 12/2011; 158(2):561-9. DOI: 10.1104/pp.111.186221
Source: PubMed

ABSTRACT

X-ray microcomputed tomography (μCT) is an invaluable tool for visualizing plant root systems within their natural soil environment noninvasively. However, variations in the x-ray attenuation values of root material and the overlap in attenuation values between roots and soil caused by water and organic materials represent major challenges to data recovery. We report the development of automatic root segmentation methods and software that view μCT data as a sequence of images through which root objects appear to move as the x-y cross sections are traversed along the z axis of the image stack. Previous approaches have employed significant levels of user interaction and/or fixed criteria to distinguish root and nonroot material. RooTrak exploits multiple, local models of root appearance, each built while tracking a specific segment, to identify new root material. It requires minimal user interaction and is able to adapt to changing root density estimates. The model-guided search for root material arising from the adoption of a visual-tracking framework makes RooTrak less sensitive to the natural ambiguity of x-ray attenuation data. We demonstrate the utility of RooTrak using μCT scans of maize (Zea mays), wheat (Triticum aestivum), and tomato (Solanum lycopersicum) grown in a range of contrasting soil textures. Our results demonstrate that RooTrak can successfully extract a range of root architectures from the surrounding soil and promises to facilitate future root phenotyping efforts.

Download full-text

Full-text

Available from: Susan Zappala
  • Source
    • "The smallest resolution (µm 3 ) can be reached with synchrotron X-ray sources, and true microtomography may then allow the detection of hair roots with a diameter of a few µm. That is for a very small part (<1 cm 3 ) of the root system, though, and with excessively large CTN datasets to analyze if expanded (Mairhofer et al., 2012). Accordingly, synchrotron-based CT scanning has very originally been established as a technology to assess the filling status of xylem vessels and detect embolisms (Lee and Kim, 2008; Choat et al., 2015), and to unravel anatomical features of the vascular system (Kim et al., 2014). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Non-medical applications of computed tomography (CT) scanning have flourished in recent years, including in Plant Science. This Perspective article on CT scanning of root systems and leaf canopies is intended to be of interest to three categories of readers: those who have not yet tried plant CT scanning, and should find inspiration for new research objectives; readers who are on the learning curve with applications—here is helpful advice for them; and researchers with greater experience—the field is evolving quickly and it is easy to miss aspects. Our conclusion is that CT scanning of roots and canopies is highly demanding in terms of technology, multidisciplinarity and big-data analysis, to name a few areas of expertise, but eventually, the reward for researchers is directly proportional!
    Preview · Article · Dec 2015 · Frontiers in Plant Science
  • Source
    • "GmbH, Wunstorf, Germany) and data was visualized using VGStudio MAX 2.2 (Volume Graphics GmbH, Heidelberg, Germany). Images of the rice root systems were separated from the images of the Turface material by segmentation and quantified using a combination of region growing segmentation techniques in VGStudioMAX and RooTrak software (Mairhofer et al. 2012 "
    [Show description] [Hide description]
    DESCRIPTION: Root system architecture is being studied more closely for improved nutrient acquisition, stress tolerance and carbon sequestration by relating the genetic material that corresponds to preferential physical features. This information can help direct plant breeders in addressing the growing concerns regarding the global demand on crops and fossil fuels. To help support this incentive comes a need to make high-throughput image-based phenotyping of plant roots, at the individual plant scale, simpler and more affordable. Our goal is to create an affordable and portable product for simple image collection, processing and management that will extend root phenotyping to institutions with limited funding (e.g., in developing countries). Thus, a new integrated system has been developed using the Raspberry Pi single-board computer. Similar to other 3D-based imaging platforms, the system utilizes a stationary camera to photograph a rotating crop root system (e.g., rice, maize or sorghum) that is suspended either in a gel or on a mesh (for hydroponics). In contrast, the new design takes advantage of powerful open-source hardware and software to reduce the system costs, simplify the imaging process, and manage the large datasets produced by the high-resolution photographs. A newly designed graphical user interface (GUI) unifies the system controls (e.g., adjusting camera and motor settings and orchestrating the motor motion with image capture), making it easier to accommodate a variety of experiments. During each imaging session, integral metadata necessary for reproducing experiment results are collected (e.g., plant type and age, growing conditions and treatments, camera settings) using hierarchical data format files. These metadata are searchable within the GUI and can be selected and extracted for further analysis. The GUI also supports an image previewer that performs limited image processing (e.g., thresholding and cropping). Root skeletonization, 3D reconstruction and trait calculation (e.g., rooting depth, rooting angle, total volume of roots) is being developed in conjunction with this project.
    Full-text · Poster · Dec 2015
    • "Recent advances in detector design have improved the sensitivity, signal : noise ratio, quantum efficiency, dynamic range, and number and size of pixels. This is especially important for root–soil studies, as the similar attenuation densities of root material and water-filled pore space has been an obstacle to progress for some time (Tracy et al., 2010 ), limiting the success of new roottracking algorithms (Mairhofer et al., 2012) to some extent. Higher-quality data would be obtained if future CT detectors could enhance the contrast between root and soil material. "

    No preview · Chapter · Oct 2015
Show more