The purpose of this work is to find an effective image segmentation method for lab-based micro-tomography (µ-CT) data of carbon fiber reinforced polymers (CFRP) with insufficient contrast-to-noise ratio. The segmentation is the first step in creating a realistic geometry (based on µ-CT) for finite element modelling of textile composites on meso-scale. Noise in X-ray imaging data of carbon/polymer composites forms a challenge for this segmentation due to the very low X-ray contrast between fiber and polymer and unclear fiber gradients. To the best of our knowledge, segmentation of µ-CT images of carbon/polymer textile composites with low resolution data (voxel size close to the fiber diameter) remains poorly documented. In this paper, we propose and evaluate different approaches for solving the segmentation problem: variational on the one hand and deep-learning-based on the other. In the author’s view, both strategies present a novel and reliable ground for the segmentation of µ-CT data of CFRP woven composites. The predictions of both approaches were evaluated against a manual segmentation of the volume, constituting our “ground truth”, which provides quantitative data on the segmentation accuracy. The highest segmentation accuracy (about 4.7% in terms of voxel-wise Dice similarity) was achieved using the deep learning approach with U-Net neural network.

Proprioceptive feedback is a critical component of voluntary movement planning and execution. Neuroprosthetic technologies aiming at restoring movement must interact with it to restore accurate motor control. Optimization and design of such technologies depends on the availability of quantitative insights into the neural dynamics of proprioceptive afferents during functional movements. However, recording proprioceptive neural activity during unconstrained movements in clinically relevant animal models presents formidable challenges. In this work, we developed a computational framework to estimate the spatiotemporal patterns of proprioceptive inputs to the cervical spinal cord during three-dimensional arm movements in monkeys. We extended a biomechanical model of the monkey arm with exvivo measurements, and combined it with models of mammalian group-Ia, Ib and II afferent fibers. We then used experimental recordings of arm kinematics and muscle activity of two monkeys performing a reaching and grasping task to estimate muscle stretches and forces with computational biomechanics. Finally, we projected the simulated proprioceptive firing rates onto the cervical spinal roots, thus obtaining spatiotemporal maps of spinal proprioceptive inputs during voluntary movements. Estimated maps show complex and markedly distinct patterns of neural activity for each of the fiber populations spanning the spinal cord rostro-caudally. Our results indicate that reproducing the proprioceptive information flow to the cervical spinal cord requires complex spatio-temporal modulation of each spinal root. Our model can support the design of neuroprosthetic technologies as well as in-silico investigations of the primate sensorimotor system.

The distribution and good spreading of adhesive resins is critical for the wood-based panels industry. Full 3D non-destructive characterization is necessary, but methods are limited due to the chemical similarities between the resins and the wood fibers. For X-ray microtomography ( $$\mu $$μ CT), the doping of the resin with a highly attenuating contrast agent is necessary to visualize the resin distribution. However, the attenuation signal remains hard to segment clearly due to partial volume effects in the image, and phase mixing in the material. To help in the identification of the doped resin, dual-energy X-ray CT (DECT) is used to exploit the contrast agent’s K-edge, based on simulations which take into account the polychromatic properties of the X-ray tube and detector response. The contrast agent’s identification with DECT is validated with elemental mapping using scanning electron microscopy combined with energy-dispersive spectroscopy (SEM-EDX) on the surface of a wood-based panel sample, using data fusion between DECT and SEM-EDX. Overall, DECT results here in the first 3D identification of doped resin inside wood fiberboards, guiding the industry’s efforts in further improving the durability of wood-based panels.

Medium-density fiberboard (MDF), a wood-based material that consists of a tight random network of wood fibers, deforms more than wood when exposed to water. For the first time, the microscopic deformations of MDF were tracked during swelling. A hygroscopic swelling setup imposing the material to deform throughout tomographic acquisition was used coupled to X-ray microtomography. An advanced reconstruction algorithm enabled reconstruction of images free of motion artefacts, and state-of-the-art digital volume correlation was applied to determine the mechanical strain fields at high resolution. Wood fiber bundles were then segmented from single fibers with deep learning using the UNet3D architecture. Combined with the strain fields, this segmentation showed that wood fiber bundles were the drivers of MDF swelling. This contrasts with the hygroscopic behavior of wood, where structured wood swells less than single fibers, which might be caused by a difference in penetration and distribution of the adhesive, in and on the wood fiber cell wall. The unique characterization of MDF’s dynamic behavior can already be used to develop manufacturing strategies to improve water resistance, therefore widening the uses of natural fiber-based materials.