21 results
Now showing1 - 10 of 21
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Enhancing 3D reconstruction accuracy of FIB tomography data using multi-voltage images and multimodal machine learningFIB-SEM tomography is a powerful technique that integrates a focused ion beam (FIB) and a scanning electron microscope (SEM) to capture high-resolution imaging data of nanostructures. This approach involves collecting in-plane SEM images and using FIB to remove material layers for imaging subsequent planes, thereby producing image stacks. However, these image stacks in FIB-SEM tomography are subject to the shine-through effect, which makes structures visible from the posterior regions of the current plane. This artifact introduces an ambiguity between image intensity and structures in the current plane, making conventional segmentation methods such as thresholding or the k-means algorithm insufficient. In this study, we propose a multimodal machine learning approach that combines intensity information obtained at different electron beam accelerating voltages to improve the three-dimensional (3D) reconstruction of nanostructures. By treating the increased shine-through effect at higher accelerating voltages as a form of additional information, the proposed method significantly improves segmentation accuracy and leads to more precise 3D reconstructions for real FIB tomography data.Publicationtype: Journal Article77 129 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Comparative analysis of ternary TiAlNb interatomic potentials: moment tensor vs. deep learning approaches(Frontiers Media S.A., 2024-10-03); ; ; ; Intermetallic titanium aluminides, leveraging the ordered γ-TiAl phase, attract increasing attention in aerospace and automotive engineering due to their favorable mechanical properties at high temperatures. Of particular interest are γ-TiAl-based alloys with a Niobium (Nb) concentration of 5–10 at.%. It is a key question how to model such ternary alloys at the atomic scale with molecular dynamics (MD) simulations to better understand (and subsequently optimize) the alloys. Here, we present a comparative analysis of ternary TiAlNb interatomic potentials developed by the moment tensor potential (MTP) and deep potential molecular dynamics (DeePMD) methods specifically for the above mentioned critical Nb concentration range. We introduce a novel dataset (TiAlNb dataset) for potential training that establishes a benchmark for the assessment of TiAlNb potentials. The potentials were evaluated through rigorous error analysis and performance metrics, alongside calculations of material properties such as elastic constants, equilibrium volume, and lattice constant. Additionally, we explore finite temperature properties including specific heat and thermal expansion with both potentials. Mechanical behaviors, such as uniaxial tension and the calculation of generalized stacking fault energy, are analyzed to determine the impact of Nb alloying in TiAl-based alloys. Our results indicate that Nb alloying generally enhances the ductility of TiAl-based alloys at the expense of reduced strength, with the notable exception of simulations using DeePMD for the γ-TiAl phase, where this trend does not apply.Publicationtype: Journal Article41 217 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , A calibration approach for elasticity estimation with medical tools(De Gruyter, 2024-09-14); ; ; Soft tissue elasticity is directly related to different stages of diseases and can be used for tissue identification during minimally invasive procedures. By palpating a tissue with a robot in a minimally invasive fashion force-displacement curves can be acquired. However, force-displacement curves strongly depend on the tool geometry which is often complex in the case of medical tools. Hence, a tool calibration procedure is desired to directly map force-displacement curves to the corresponding tissue elasticity.We present an experimental setup for calibrating medical tools with a robot. First, we propose to estimate the elasticity of gelatin phantoms by spherical indentation with a state-of-The-Art contact model. We estimate force-displacement curves for different gelatin elasticities and temperatures. Our experiments demonstrate that gelatin elasticity is highly dependent on temperature, which can lead to an elasticity offset if not considered. Second, we propose to use a more complex material model, e.g., a neural network, that can be trained with the determined elasticities. Considering the temperature of the gelatin sample we can represent different elasticities per phantom and thereby increase our training data.We report elasticity values ranging from 10 to 40 kPa for a 10% gelatin phantom, depending on temperature.Publicationtype: Journal Article16 55 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Viscoelastic constitutive artificial neural networks (vCANNs) – A framework for data-driven anisotropic nonlinear finite viscoelasticityThe constitutive behavior of polymeric materials is often modeled by finite linear viscoelastic (FLV) or quasi-linear viscoelastic (QLV) models. These popular models are simplifications that typically cannot accurately capture the nonlinear viscoelastic behavior of materials. For example, the success of attempts to capture strain (rate)-dependent behavior has been limited so far. To overcome this problem, we introduce viscoelastic Constitutive Artificial Neural Networks (vCANNs), a novel physics-informed machine learning framework for anisotropic nonlinear viscoelasticity at finite strains. vCANNs rely on the concept of generalized Maxwell models enhanced with nonlinear strain (rate)-dependent properties represented by neural networks. The flexibility of vCANNs enables them to automatically identify accurate and sparse constitutive models of a broad range of materials. To test vCANNs, we trained them on stress-strain data from Polyvinyl Butyral, the electro-active polymers VHB 4910 and 4905, and a biological tissue, the rectus abdominis muscle. Different loading conditions were considered, including relaxation tests, cyclic tension-compression tests, and blast loads. We demonstrate that vCANNs can learn to capture the behavior of all these materials accurately and computationally efficiently without human guidance. Our source code is available at https://github.com/ConstitutiveANN/vCANN.Publicationtype: Journal ArticleScopus© Citations 4 38 220 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Role of slice thickness quantification in the 3D reconstruction of FIB tomography data of nanoporous materials(Elsevier, 2024-02); ; ; ; In focused ion beam (FIB) tomography, a combination of FIB with a scanning electron microscope (SEM) is used for collecting a series of planar images of the microstructure of nanoporous materials. These planar images serve as the basis for reconstructing the three-dimensional microstructure through segmentation algorithms. However, the assumption of a constant distance between consecutively imaged sections is generally invalid due to random variations in the FIB milling process. This variation complicates the accurate reconstruction of the three-dimensional microstructure. Using synthetic FIB tomography data, we present an algorithm that repositions slices according to their actual thickness and interpolates the results using machine learning-based methods. We applied our algorithm to real datasets, comparing two standard approaches of microstructure reconstruction: on-the-fly via image processing and ruler-based via sample structuring. Our findings indicate that the ruler-based method, combined with our novel slice repositioning and interpolation algorithm, exhibits superior performance in reconstructing the microstructure.Publicationtype: Journal ArticleScopus© Citations 2 63 128 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Searching the chemical space for effective magnesium dissolution modulators: a deep learning approach using sparse features(Nature Publishing Group, 2023-12-01); ; ; ; Small organic molecules can alter the degradation rates of the magnesium alloy ZE41. However, identifying suitable candidate compounds from the vast chemical space requires sophisticated tools. The information contained in only a few molecular descriptors derived from recursive feature elimination was previously shown to hold the potential for determining such candidates using deep neural networks. We evaluate the capability of these networks to generalise by blind testing them on 15 randomly selected, completely unseen compounds. We find that their generalisation ability is still somewhat limited, most likely due to the relatively small amount of available training data. However, we demonstrate that our approach is scalable; meaning deficiencies caused by data limitations can presumably be overcome as the data availability increases. Finally, we illustrate the influence and importance of well-chosen descriptors towards the predictive power of deep neural networks.Publicationtype: Journal Article36 40 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , 23 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Predicting and understanding arterial elasticity from key microstructural features by bidirectional deep learningMicrostructural features and mechanical properties are closely related in all soft biological tissues. Both yet exhibit considerable inter-individual differences and are affected by factors such as aging and disease and its progression. Histological analysis, modern in situ imaging, and biomechanical testing have deepened our understanding of these complex interrelations, yet two key questions remain: (1) Given the specific microstructure, can one predict the macroscopic mechanical properties without mechanical testing? (2) Can one quantify individual contributions of the different microstructural features to the macroscopic mechanical properties in an automated, systematic and largely unbiased way? Here we propose a bidirectional deep learning architecture to address these two questions. Our architecture uses data from standard histological analyses, two-photon microscopy and biaxial biomechanical testing. Its capabilities are demonstrated by predicting with high accuracy (R2=0.92) the evolving mechanical properties of the murine aorta during maturation and aging. Moreover, our architecture reveals that the extracellular matrix composition and organization are the most prominent factors governing the macroscopic mechanical properties of the tissues studied herein. Statement of significance:Publicationtype: Journal ArticleScopus© Citations 7 75 302 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Ultrasound shear wave elasticity imaging with spatio-temporal deep learning(IEEE, 2022-04-19); ; ; ; Ultrasound shear wave elasticity imaging is a valuable tool for quantifying the elastic properties of tissue. Typically, the shear wave velocity is derived and mapped to an elasticity value, which neglects information such as the shape of the propagating shear wave or push sequence characteristics. We present 3D spatio-temporal CNNs for fast local elasticity estimation from ultrasound data. This approach is based on retrieving elastic properties from shear wave propagation within small local regions. A large training data set is acquired with a robot from homogeneous gelatin phantoms ranging from 17.42 kPa to 126.05 kPa with various push locations. The results show that our approach can estimate elastic properties on a pixelwise basis with a mean absolute error of 5.01(437) kPa. Furthermore, we estimate local elasticity independent of the push location and can even perform accurate estimates inside the push region. For phantoms with embedded inclusions, we report a 53.93% lower MAE (7.50 kPa) and on the background of 85.24% (1.64 kPa) compared to a conventional shear wave method. Overall, our method offers fast local estimations of elastic properties with small spatio-temporal window sizes.Publicationtype: Journal ArticleScopus© Citations 5 88 231 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Training deep neural networks to reconstruct nanoporous structures from FIB tomography images using synthetic training data(Frontiers Media S.A., 2022-02-28); ; ; ; Focused ion beam (FIB) tomography is a destructive technique used to collect three-dimensional (3D) structural information at a resolution of a few nanometers. For FIB tomography, a material sample is degraded by layer-wise milling. After each layer, the current surface is imaged by a scanning electron microscope (SEM), providing a consecutive series of cross-sections of the three-dimensional material sample. Especially for nanoporous materials, the reconstruction of the 3D microstructure of the material, from the information collected during FIB tomography, is impaired by the so-called shine-through effect. This effect prevents a unique mapping between voxel intensity values and material phase (e.g., solid or void). It often substantially reduces the accuracy of conventional methods for image segmentation. Here we demonstrate how machine learning can be used to tackle this problem. A bottleneck in doing so is the availability of sufficient training data. To overcome this problem, we present a novel approach to generate synthetic training data in the form of FIB-SEM images generated by Monte Carlo simulations. Based on this approach, we compare the performance of different machine learning architectures for segmenting FIB tomography data of nanoporous materials. We demonstrate that two-dimensional (2D) convolutional neural network (CNN) architectures processing a group of adjacent slices as input data as well as 3D CNN perform best and can enhance the segmentation performance significantly.Publicationtype: Journal ArticleScopus© Citations 5 156 906