Machine learning denoising of high-resolution X-ray nanotomography data

A Nanoscale View of the Structure and Deformation Mechanism of Mineralized Shark Vertebral Cartilage

Swimming kinematics and macroscale mechanical testing have shown that the vertebral column of sharks acts as a biological spring, storing and releasing energy during locomotion. Using synchrotron X-ray nanotomography and deep-learning image segmentation, we studied the ultrastructure and deformation mechanism of mineralized shark vertebrae from Carcharhinus limbatus (Blacktip shark). The vertebral centrum con regions: the corpus calcareum, a hypermineralized double cone, and the intermediale, blocks of mineralized cartilage interspersed by unmineralized arches. At the micron scale, mineralized cartilage has previously been described as a 3D network of interconnected mineral plates that vary in thickness and spacing. The corpus calcareum consists of stacked, interconnected, curved mineralized planes permeated by a network of organic occlusions. The mineral network in the intermedialia resembles trabecular bone, including thicker struts in the direction opposite to the predominant biological strain. We characterized collagenous fiber elements winding around lacunar spaces in the intermedialia, and we hypothesize the swirling arrangement and elasticity of the fibers to be distributing stress. With little permanent deformation detected in mineralized structures, it is likely that the soft organic matrix is crucial for absorbing energy through deformation, irreversible damage, and viscoelastic behavior. In the corpus calcareum, cracks typically terminate toward thick struts along the mineral planes, resembling the microscale crack deflection and arrest mechanism found in other staggered biocomposites, such as nacre or bone. Using transmission electron microscopy (TEM), we observed preferentially oriented, needlelike bioapatite crystallites and d-band patterns of collagen type-II fibrils resulting from intrafibrillar mineralization.

Self-supervised denoising of grating-based phase-contrast computed tomography

In the last decade, grating-based phase-contrast computed tomography (gbPC-CT) has received growing interest. It provides additional information about the refractive index decrement in the sample. This signal shows an increased soft-tissue contrast. However, the resolution dependence of the signal poses a challenge: its contrast enhancement is overcompensated by the low resolution in low-dose applications such as clinical computed tomography. As a result, the implementation of gbPC-CT is currently tied to a higher dose. To reduce the dose, we introduce the self-supervised deep learning network Noise2Inverse into the field of gbPC-CT. We evaluate the behavior of the Noise2Inverse parameters on the phase-contrast results. Afterward, we compare its results with other denoising methods, namely the Statistical Iterative Reconstruction, Block Matching 3D, and Patchwise Phase Retrieval. In the example of Noise2Inverse, we show that deep learning networks can deliver superior denoising results with respect to the investigated image quality metrics. Their application allows to increase the resolution while maintaining the dose. At higher resolutions, gbPC-CT can naturally deliver higher contrast than conventional absorption-based CT. Therefore, the application of machine learning-based denoisers shifts the dose-normalized image quality in favor of gbPC-CT, bringing it one step closer to medical application.

Using convolutional neural network denoising to reduce ambiguity in X-ray coherent diffraction imaging

The inherent ambiguity in reconstructed images from coherent diffraction imaging (CDI) poses an intrinsic challenge, as images derived from the same dataset under varying initial conditions often display inconsistencies. This study introduces a method that employs the Noise2Noise approach combined with neural networks to effectively mitigate these ambiguities. We applied this methodology to hundreds of ambiguous reconstructed images retrieved from a single diffraction pattern using a conventional retrieval algorithm. Our results demonstrate that ambiguous features in these reconstructions are effectively treated as inter-reconstruction noise and are significantly reduced. The post-Noise2Noise treated images closely approximate the average and singular value decomposition analysis of various reconstructions, providing consistent and reliable reconstructions.

Speckle tracking phase-contrast computed tomography at an inverse Compton X-ray source

Speckle-based X-ray imaging (SBI) is a phase-contrast method developed at and for highly coherent X-ray sources, such as synchrotrons, to increase the contrast of weakly absorbing objects. Consequently, it complements the conventional attenuation-based X-ray imaging. Meanwhile, attempts to establish SBI at less coherent laboratory sources have been performed, ranging from liquid metal-jet X-ray sources to microfocus X-ray tubes. However, their lack of coherence results in interference fringes not being resolved. Therefore, algorithms were developed which neglect the interference effects. Here, we demonstrate phase-contrast computed tomography employing SBI in a laboratory-setting with an inverse Compton X-ray source. In this context, we investigate and compare also the performance of the at synchrotron conventionally used phase-retrieval algorithms for SBI, unified modulated pattern analysis (UMPA) with a phase-retrieval method developed for low coherence systems (LCS). We successfully retrieve a full computed tomography in a phantom as well as in biological specimens, such as larvae of the greater wax moth (Galleria mellonella), a model system for studies of pathogens and infections. In this context, we additionally demonstrate quantitative phase-contrast computed tomography using SBI at a low coherent set-up.

Dual-beam X-ray nano-holotomography

Nanotomography with hard X-rays is a widely used technique for high-resolution imaging, providing insights into the structure and composition of various materials. In recent years, tomographic approaches based on simultaneous illuminations of the same sample region from different angles by multiple beams have been developed at micrometre image resolution. Transferring these techniques to the nanoscale is challenging due to the loss in photon flux by focusing the X-ray beam. We present an approach for multi-beam nanotomography using a dual-beam Fresnel zone plate (dFZP) in a near-field holography setup. The dFZP generates two nano-focused beams that overlap in the sample plane, enabling the simultaneous acquisition of two projections from slightly different angles. This first proof-of-principle implementation of the dual-beam setup allows for the efficient removal of ring artifacts and noise using machine-learning approaches. The results open new possibilities for full-field multi-beam nanotomography and pave the way for future advancements in fast holotomography and artifact-reduction techniques.

Deep learning to overcome Zernike phase-contrast nanoCT artifacts for automated micro-nano porosity segmentation in bone

Bone material contains a hierarchical network of micro- and nano-cavities and channels, known as the lacuna-canalicular network (LCN), that is thought to play an important role in mechanobiology and turnover. The LCN comprises micrometer-sized lacunae, voids that house osteocytes, and submicrometer-sized canaliculi that connect bone cells. Characterization of this network in three dimensions is crucial for many bone studies. To quantify X-ray Zernike phase-contrast nanotomography data, deep learning is used to isolate and assess porosity in artifact-laden tomographies of zebrafish bones. A technical solution is proposed to overcome the halo and shade-off domains in order to reliably obtain the distribution and morphology of the LCN in the tomographic data. Convolutional neural network (CNN) models are utilized with increasing numbers of images, repeatedly validated by `error loss' and `accuracy' metrics. U-Net and Sensor3D CNN models were trained on data obtained from two different synchrotron Zernike phase-contrast transmission X-ray microscopes, the ANATOMIX beamline at SOLEIL (Paris, France) and the P05 beamline at PETRA III (Hamburg, Germany). The Sensor3D CNN model with a smaller batch size of 32 and a training data size of 70 images showed the best performance (accuracy 0.983 and error loss 0.032). The analysis procedures, validated by comparison with human-identified ground-truth images, correctly identified the voids within the bone matrix. This proposed approach may have further application to classify structures in volumetric images that contain non-linear artifacts that degrade image quality and hinder feature identification.

On the material dependency of peri-implant morphology and stability in healing bone

The microstructural architecture of remodeled bone in the peri-implant region of screw implants plays a vital role in the distribution of strain energy and implant stability. We present a study in which screw implants made from titanium, polyetheretherketone and biodegradable magnesium-gadolinium alloys were implanted into rat tibia and subjected to a push-out test four, eight and twelve weeks after implantation. Screws were 4 mm in length and with an M2 thread. The loading experiment was accompanied by simultaneous three-dimensional imaging using synchrotron-radiation microcomputed tomography at 5 μm resolution. Bone deformation and strains were tracked by applying optical flow-based digital volume correlation to the recorded image sequences. Implant stabilities measured for screws of biodegradable alloys were comparable to pins whereas non-degradable biomaterials experienced additional mechanical stabilization. Peri-implant bone morphology and strain transfer from the loaded implant site depended heavily on the biomaterial utilized. Titanium implants stimulated rapid callus formation displaying a consistent monomodal strain profile whereas the bone volume fraction in the vicinity of magnesium-gadolinium alloys exhibited a minimum close to the interface of the implant and less ordered strain transfer. Correlations in our data suggest that implant stability benefits from disparate bone morphological properties depending on the biomaterial utilized. This leaves the choice of biomaterial as situational depending on local tissue properties.

Machine learning-assisted data filtering and QSAR models for prediction of chemical acute toxicity on rat and mouse

Machine learning (ML) methods provide a new opportunity to build quantitative structure-activity relationship (QSAR) models for predicting chemicals' toxicity based on large toxicity data sets, but they are limited in insufficient model robustness due to poor data set quality for chemicals with certain structures. To address this issue and improve model robustness, we built a large data set on rat oral acute toxicity for thousands of chemicals, then used ML to filter chemicals favorable for regression models (CFRM). In comparison to chemicals not favorable for regression models (CNRM), CFRM accounted for 67% of chemicals in the original data set, and had a higher structural similarity and a smaller toxicity distribution in 2-4 log₁₀ (mg/kg). The performance of established regression models for CFRM was greatly improved, with root-mean-square deviations (RMSE) in the range of 0.45-0.48 log₁₀ (mg/kg). Classification models were built for CNRM using all chemicals in the original data set, and the area under receiver operating characteristic (AUROC) reached 0.75-0.76. The proposed strategy was successfully applied to a mouse oral acute data set, yielding RMSE and AUROC in the range of 0.36-0.38 log₁₀ (mg/kg) and 0.79, respectively.

Hard X-ray full-field nanoimaging using a direct photon-counting detector

Full-field X-ray nanoimaging is a widely used tool in a broad range of scientific areas. In particular, for low-absorbing biological or medical samples, phase contrast methods have to be considered. Three well established phase contrast methods at the nanoscale are transmission X-ray microscopy with Zernike phase contrast, near-field holography and near-field ptychography. The high spatial resolution, however, often comes with the drawback of a lower signal-to-noise ratio and significantly longer scan times, compared with microimaging. In order to tackle these challenges a single-photon-counting detector has been implemented at the nanoimaging endstation of the beamline P05 at PETRA III (DESY, Hamburg) operated by Helmholtz-Zentrum Hereon. Thanks to the long sample-to-detector distance available, spatial resolutions of below 100 nm were reached in all three presented nanoimaging techniques. This work shows that a single-photon-counting detector in combination with a long sample-to-detector distance allows one to increase the time resolution for in situ nanoimaging, while keeping a high signal-to-noise level.

Deploying Machine Learning Based Segmentation for Scientific Imaging Analysis at Synchrotron Facilities

Scientific user facilities present a unique set of challenges for image processing due to the large volume of data generated from experiments and simulations. Furthermore, developing and implementing algorithms for real-time processing and analysis while correcting for any artifacts or distortions in images remains a complex task, given the computational requirements of the processing algorithms. In a collaborative effort across multiple Department of Energy national laboratories, the "MLExchange" project is focused on addressing these challenges. MLExchange is a Machine Learning framework deploying interactive web interfaces to enhance and accelerate data analysis. The platform allows users to easily upload, visualize, label, and train networks. The resulting models can be deployed on real data while both results and models could be shared with the scientists. The MLExchange web-based application for image segmentation allows for training, testing, and evaluating multiple machine learning models on hand-labeled tomography data. This environment provides users with an intuitive interface for segmenting images using a variety of machine learning algorithms and deep-learning neural networks. Additionally, these tools have the potential to overcome limitations in traditional image segmentation techniques, particularly for complex and low-contrast images.

Performance Evaluation of Deep Neural Network Model for Coherent X-ray Imaging

We present a supervised deep neural network model for phase retrieval of coherent X-ray imaging and evaluate the performance. A supervised deep-learning-based approach requires a large amount of pre-training datasets. In most proposed models, the various experimental uncertainties are not considered when the input dataset, corresponding to the diffraction image in reciprocal space, is generated. We explore the performance of the deep neural network model, which is trained with an ideal quality of dataset, when it faces real-like corrupted diffraction images. We focus on three aspects of data qualities such as a detection dynamic range, a degree of coherence and noise level. The investigation shows that the deep neural network model is robust to a limited dynamic range and partially coherent X-ray illumination in comparison to the traditional phase retrieval, although it is more sensitive to the noise than the iteration-based method. This study suggests a baseline capability of the supervised deep neural network model for coherent X-ray imaging in preparation for the deployment to the laboratory where diffraction images are acquired.