Explore the vast range of titles available.

Automated, reliable, and objective microstructure inference from micrographs is essential for a comprehensive understanding of process-microstructure-property relations and tailored materials development. However, such inference, with the increasing complexity of microstructures, requires advanced segmentation methodologies. While deep learning offers new opportunities, an intuition about the required data quality/quantity and a methodological guideline for microstructure quantification is still missing. This, along with deep learning’s seemingly intransparent decision-making process, hampers its breakthrough in this field. We apply a multidisciplinary deep learning approach, devoting equal attention to specimen preparation and imaging, and train distinct U-Net architectures with 30–50 micrographs of different imaging modalities and electron backscatter diffraction-informed annotations. On the challenging task of lath-bainite segmentation in complex-phase steel, we achieve accuracies of 90% rivaling expert segmentations. Further, we discuss the impact of image context, pre-training with domain-extrinsic data, and data augmentation. Network visualization techniques demonstrate plausible model decisions based on grain boundary morphology. Segmentation and classification of microstructures are required by quality control and materials development. The authors apply deep learning for the segmentation of complex phase steel microstructures, providing a bridge between experimental and computational methods for materials analysis.

The local prediction of fatigue damage within polycrystals in a high-cycle fatigue setting is a long-lasting and challenging task. It requires identifying grains tending to accumulate plastic deformation under cyclic loading. We address this task by transcribing ferritic steel microtexture and damage maps from experiments into a microstructure graph. Here, grains constitute graph nodes connected by edges whenever grains share a common boundary. Fatigue loading causes some grains to develop slip markings, which can evolve into microcracks and lead to failure. This data set enables applying graph neural network variants on the task of binary grain-wise damage classification. The objective is to identify suitable data representations and models with an appropriate inductive bias to learn the underlying damage formation causes. Here, graph convolutional networks yielded the best performance with a balanced accuracy of 0.72 and a F 1 -score of 0.34, outperforming phenomenological crystal plasticity (+ 68%) and conventional machine learning (+ 17%) models by large margins. Further, we present an interpretability analysis that highlights the grains along with features that are considered important by the graph model for the prediction of fatigue damage initiation, thus demonstrating the potential of such techniques to reveal underlying mechanisms and microstructural driving forces in critical grain ensembles.

The COVID-19 pandemic has accelerated methods to facilitate contactless evaluation of patients in hospital settings. By minimizing in-person contact with individuals who may have COVID-19, healthcare workers can prevent disease transmission and conserve personal protective equipment. Obtaining vital signs is a ubiquitous task that is commonly done in person by healthcare workers. To eliminate the need for in-person contact for vital sign measurement in the hospital setting, we developed Dr. Spot, a mobile quadruped robotic system. The system includes IR and RGB cameras for vital sign monitoring and a tablet computer for face-to-face medical interviewing. Dr. Spot is teleoperated by trained clinical staff to simultaneously measure the skin temperature, respiratory rate, and heart rate while maintaining social distancing from patients and without removing their mask. To enable accurate, contactless measurements on a mobile system without a static black body as reference, we propose novel methods for skin temperature compensation and respiratory rate measurement at various distances between the subject and the cameras, up to 5 m. Without compensation, the skin temperature MAE is 1.3°C. Using the proposed compensation method, the skin temperature MAE is reduced to 0.3°C. The respiratory rate method can provide continuous monitoring with a MAE of 1.6 BPM in 30 s or rapid screening with a MAE of 2.1 BPM in 10 s. For the heart rate estimation, our system is able to achieve a MAE less than 8 BPM in 10 s measured in arbitrary indoor light conditions at any distance below 2 m.

While large language models learn sound statistical representations of the language and information therein, ontologies are symbolic knowledge representations that can complement the former ideally. Research at this critical intersection relies on datasets that intertwine ontologies and text corpora to enable training and comprehensive benchmarking of neurosymbolic models. We present the MaterioMiner dataset and the linked materials mechanics ontology where ontological concepts from the mechanics of materials domain are associated with textual entities within the literature corpus. Another distinctive feature of the dataset is its eminently fine-grained annotation. Specifically, 179 distinct classes are manually annotated by three raters within four publications, amounting to 2191 entities that were annotated and curated. Conceptual work is presented for the symbolic representation of causal composition-process-microstructure-property relationships. We explore the annotation consistency between the three raters and perform fine-tuning of pre-trained language models to showcase the feasibility of training named entity recognition models. Reusing the dataset can foster training and benchmarking of materials language models, automated ontology construction, and knowledge graph generation from textual data.

Extracting structured and semantically annotated materials information from unstructured scientific literature is a crucial step toward constructing machine-interpretable knowledge graphs and accelerating data-driven materials research. This is especially important in materials science, which is adversely affected by data scarcity. Data scarcity further motivates employing solutions such as foundation language models for extracting information which can in principle address several subtasks of the information extraction problem in a range of domains without the need of generating costly large-scale annotated datasets for each downstream task. However, foundation language models struggle with tasks like Named Entity Recognition (NER) due to domain-specific terminologies, fine-grained entities, and semantic ambiguity. The issue is even more pronounced when entities must map directly to pre-existing domain ontologies. This work aims to assess whether foundation large language models (LLMs) can successfully perform ontology-conformal NER in the materials mechanics and fatigue domain. Specifically, we present a comparative evaluation of in-context learning (ICL) with foundation models such as GPT-4 against fine-tuned task-specific language models, including MatSciBERT and DeBERTa. The study is performed on two materials fatigue datasets, which contain annotations at a comparatively fine-grained level adhering to the class definitions of a formal ontology to ensure semantic alignment and cross-dataset interoperability. Both datasets cover adjacent domains to assess how well both NER methodologies generalize when presented with typical domain shifts. Task-specific models are shown to significantly outperform general foundation models on an ontology-constrained NER. Our findings reveal a strong dependence on the quality of few-shot demonstrations in ICL to handle domain-shift. The study also highlights the significance of domain-specific pre-training by comparing task-specific models that differ primarily in their pre-training corpus.

Materials’ microstructures are signatures of their alloying composition and processing history. Automated, quantitative analyses of microstructural constituents were lately accomplished through deep learning approaches. However, their shortcomings are poor data efficiency and domain generalizability across data sets, inherently conflicting the expenses associated with annotating data through experts, and extensive materials diversity. To tackle both, we propose to apply a sub-class of transfer learning methods called unsupervised domain adaptation (UDA). UDA addresses the task of finding domain-invariant features when supplied with annotated source data and unannotated target data, such that performance on the latter is optimized. Exemplarily, this study is conducted on a lath-shaped bainite segmentation task in complex phase steel micrographs. Domains to bridge are selected to be different metallographic specimen preparations and distinct imaging modalities. We show that a state-of-the-art UDA approach substantially fosters the transfer between the investigated domains, underlining this technique’s potential to cope with materials variance.

This paper proposes a robust methodology for integrating process-specific data and domain expert knowledge into linked knowledge graphs. These graphs utilize an ontology that provides a standardized vocabulary for material science and facilitates the creation of semantic models for various processes along the digital process chain. A generic template for structuring processes is proposed, simplifying subsequent data retrieval. The templates of specific processes are designed collaboratively by domain and ontology experts, aided by a proposed interview template that bridges the knowledge gap. Following the digitalization of material data through semantic modeling, machine-readable data with contextual metadata is stored in a graph database, which can be efficiently queried using the SPARQL language, enabling seamless integration into data pipelines. To demonstrate this approach, a knowledge graph is developed to represent the process chain of AlSi10Mg objects manufactured via permanent mold casting, capturing their complete history from the initial manufacturing step to final non-destructive testing and mechanical characterization. This methodology enhances data interoperability and accessibility while providing context-rich data for training AI models, potentially accelerating new knowledge discovery in material science.

The objective of this study was to assess the feasibility of using a poly(vinylidene fluoride) (PVDF) membrane modified with cellulose/nanostructures as a separation technique for the removal of poly(vinyl alcohol)(PVA)/reactive dyes from synthetic textile wastewater. The goal was to recycle PVA/reactive dye yellow 145 for reuse in the industry while simultaneously reclaiming water for reuse. To achieve this, the study aimed to evaluate the influence of SnO 2 /ZnO nanostructures on the polymer mixture, examining their impact on permeation and rejection of PVA/reactive dye. Additionally, the study investigated the antifouling properties of PVDF, both in the presence and absence of electrospun cellulose nanofibers. Chemical analysis techniques, including SEM, EDS, FTIR, mechanical strength testing, contact angle measurement, AFM, and determination of molecular weight cutoff (MWCO), were employed to assess the synthesized membranes. The MWCO results indicated a decrease in pore size after surface modification with electrospun cellulose acetate (CA), with the modified membrane (M2-Mod) showing a reduced MWCO of 6700 Da compared to the unmodified membrane’s MWCO of 13,980 Da. Furthermore, the study aimed to identify the optimal polymeric nanocomposite of PVDF with nano-SnO 2 or ZnO, along with electrospun cellulose nanofibers, to enhance %PVA and %dye rejection while improving membrane productivity and fouling resistance. The formulation containing a mixture of SnO 2 and ZnO, in the presence of electrospun CA, demonstrated superior performance, achieving 98% PVA rejection, 95% reactive dye rejection, and a stable flux of 20 LMH, with a normalized flux of 92%. Overall, it can be concluded that the optimized modified membrane formulation (M2-Mod) exhibited excellent antifouling behavior, holding significant potential for promoting circular economy and sustainability in textile wastewater treatment.

ObjectivesMask adherence continues to be a critical public health measure to prevent transmission of aerosol pathogens, such as SARS-CoV-2. We aimed to develop and deploy a computer vision algorithm to provide real-time feedback of mask wearing among staff in a hospital.DesignSingle-site, observational cohort study.SettingAn urban, academic hospital in Boston, Massachusetts, USA.ParticipantsWe enrolled adult hospital staff entering the hospital at a key ingress point.InterventionsConsenting participants entering the hospital were invited to experience the computer vision mask detection system. Key aspects of the detection algorithm and feedback were described to participants, who then completed a quantitative assessment to understand their perceptions and acceptance of interacting with the system to detect their mask adherence.Outcome measuresPrimary outcomes were willingness to interact with the mask system, and the degree of comfort participants felt in interacting with a public facing computer vision mask algorithm.ResultsOne hundred and eleven participants with mean age 40 (SD15.5) were enrolled in the study. Males (47.7%) and females (52.3%) were equally represented, and the majority identified as white (N=54, 49%). Most participants (N=97, 87.3%) reported acceptance of the system and most participants (N=84, 75.7%) were accepting of deployment of the system to reinforce mask adherence in public places. One third of participants (N=36) felt that a public facing computer vision system would be an intrusion into personal privacy.Public-facing computer vision software to detect and provide feedback around mask adherence may be acceptable in the hospital setting. Similar systems may be considered for deployment in locations where mask adherence is important.