Despite increasing demand, imaging the internal structure of plant organs or tissues without the use of transgenic lines expressing fluorescent proteins remains a challenge. Techniques such as magnetic resonance imaging, optical projection tomography or X-ray absorption tomography have been used with various success, depending on the size and physical properties of the biological material. X-ray in-line phase tomography was applied for the imaging of internal structures of maize seeds at early stages of development, when the cells are metabolically fully active and water is the main cell content. This 3D imaging technique with histology-like spatial resolution is demonstrated to reveal the anatomy of seed compartments with unequalled contrast by comparison with X-ray absorption tomography. An associated image processing pipeline allowed to quantitatively segment in 3D the four compartments of the seed (embryo, endosperm, nucellus and pericarp) from 7 to 21 days after pollination. This work constitutes an innovative quantitative use of X-ray in-line phase tomography as a non-destructive fast method to perform virtual histology and extends the developmental stages accessible by this technique which had previously been applied in seed biology to more mature samples.

We review a set of recent multiscale imaging techniques, producing high-resolution images of interest for plant sciences. These techniques are promising because they match the multiscale structure of plants. However, the use of such high-resolution images is challenging in the perspective of their application to high-throughput phenotyping on large populations of plants, because of the memory cost for their data storage and the computational cost for their processing to extract information. We discuss how this renews the interest for multiscale image processing tools such as wavelets, fractals and recent variants to analyse such high-resolution images.

One of the most challenging computer vision problem in plant sciences is the segmentation of root and soil from X-ray tomography. So far, this has been addressed from classical image analysis methods. In this paper, we address this root/soil segmentation problem from X-ray tomography using a new deep learning classification technique. The robustness of this technique, tested for the first time on this plant science problem, is established with root/soil presenting a very low contrast in X-ray tomography. We also demonstrate the possibility to segment efficiently root from soil while learning on purely synthetic soil and root.

Background and Purpose—: This study examines whether lesion shape documented on magnetic resonance diffusion-weighted imaging during acute stroke improves the prediction of the final infarct volume compared with lesion volume only. Methods—: Diffusion-weighted imaging data and clinical information were retrospectively reviewed in 110 consecutive patients who underwent (n=67) or not (n=43) thrombolytic therapy for acute ischemic stroke. Three-dimensional shape analysis was performed on admission diffusion-weighted imaging data and 5 shape descriptors were developed. Final infarct volume was measured on T2-fluid-attenuated inversion recovery imaging data performed 30 days after stroke. Results—: Shape analysis of acute ischemic lesion and more specifically the ratio of the bounding box volume to the lesion volume before thrombolytic treatment improved the prediction of the final infarct for patients undergoing thrombolysis ( R ² =0.86 in model with volume; R ² =0.98 in model with volume and shape). Conclusions—: Our findings suggest that lesion shape contains important predictive information and reflects important environmental factors that might determine the progression of ischemia from the core.