The open archive for STFC research publications

Full Record Details

Persistent URL http://purl.org/net/epubs/work/54062923
Record Status Checked
Record Id 54062923
Title Development of correlative neutron and X-ray computed tomography to study fluid dynamics and structural deformation at the micro-scale in plant and soil systems
Abstract The interactions between plant roots and soil are an area of active research, particularly in terms of water and nutrient uptake. Since non-invasive, in vivo studies are required, tomographic imaging appears an obvious method to use, but no one imaging modality is well suited to capture the complete system. X-ray imaging gives clear insight into soil structure and composition. However, water is very weakly attenuating to X-rays, and biological matter also displays poor contrast. Neutron imaging presents a complementary view where water and biological matter are better distinguished, but the soil minerals are imaged with inferior contrast and resolution in comparison to equivalent X-rays scans. This work aims to develop robust methods for complementary X-ray/neutron tomographic imaging of plant root samples. These should lead to new insight into water and nutrient transport in soil. The primary challenges of this project are: to develop experiments that will meet the requirements of both imaging modalities and the biological requirements of the plant samples and to develop ways to register a pair of reconstructed volume images of samples that have been produced at entirely separate facilities. This work investigates the use of fiducial markers for point-based registration concerning the material, number and distribution of markers to address the registration challenge, first with simulation and then experimentally with plant samples imaged using neutrons and X-rays. The neutron scans were collected at the IMAT instrument at ISIS Neutron and Muon Source and the X-ray scans both at X-ray Imaging Centre at the University of Southampton and the I12 beamline at Diamond Light Source. A marker segmentation algorithm designed to automate the registration process is presented and evaluated, as are methods for combining the registered data from the two modalities to optimise the technique and facilitate segmentation, quantification and further analysis.
Organisation ISIS , STFC , ISIS-IMAT
Funding Information
Related Research Object(s):
Licence Information:
Language English (EN)
Type Details URI(s) Local file(s) Year
Thesis PhD, University of Southampton, 2022. https://eprints.s…/467525/1/Thesis.pdf 2022