Search

Materials design is a critical driver of innovation, yet overlooking the technological, economic, and environmental risks inherent in materials and their supply chains can lead to unsustainable and risk-prone solutions. To address this, we present a novel risk-aware design approach that integrates Supply-Chain Aware Design Strategies into the materials development process. This approach leverages existing language models and text analysis to develop a specialized model for predicting materials feedstock supply risk indices. To efficiently navigate the multi-objective, multi-constraint design space, we employ Batch Bayesian Optimization (BBO), enabling the identification of Pareto-optimal high entropy alloys (HEAs) that balance performance objectives with minimized supply risk. A case study using the MoNbTiVW system demonstrates the efficacy of our approach in four scenarios, highlighting the significant impact of incorporating supply risk into the design process. By optimizing for both performance and supply risk, we ensure that the developed alloys are not only high-performing but also sustainable and economically viable. This integrated approach represents a critical step towards a future where materials discovery and design seamlessly consider sustainability, supply chain dynamics, and comprehensive life cycle analysis., Comment: 26 pages, 11 figures, submitted to Materialia

Algorithmic materials discovery is a multi-disciplinary domain that integrates insights from specialists in alloy design, synthesis, characterization, experimental methodologies, computational modeling, and optimization. Central to this effort is a robust data management system paired with an interactive work platform. This platform should empower users to not only access others data but also integrate their analyses, paving the way for sophisticated data pipelines. To realize this vision, there is a need for an integrative collaboration platform, streamlined data sharing and analysis tools, and efficient communication channels. Such a collaborative mechanism should transcend geographical barriers, facilitating remote interaction and fostering a challenge-response dynamic. In this paper, we present our ongoing efforts in addressing the critical challenges related to an accelerated Materials Discovery Framework as a part of the High-Throughput Materials Discovery for Extreme Conditions Initiative. Our BIRDSHOT Center has successfully harnessed various tools and strategies, including the utilization of cloud-based storage, a standardized sample naming convention, a structured file system, the implementation of sample travelers, a robust sample tracking method, and the incorporation of knowledge graphs for efficient data management. Additionally, we present the development of a data collection platform, reinforcing seamless collaboration among our team members. In summary, this paper provides an illustration and insight into the various elements of an efficient and effective workflow within an accelerated materials discovery framework while highlighting the dynamic and adaptable nature of the data management tools and sharing platforms., Comment: 28 pages, 9 figures, 2 tables, with appendix that has 8 pages, accepted for publication at IMMI

Gene duplication is a fundamental evolutionary mechanism that contributes to biological complexity and diversity (Fortna et al., 2004). Traditionally, research has focused on the duplication of gene sequences (Zhang, 1914). However, evidence suggests that the duplication of regulatory elements may also play a significant role in the evolution of genomic functions (Teichmann and Babu, 2004; Hallin and Landry, 2019). In this work, the evolution of regulatory relationships belonging to gene-specific-substructures in a GRN are modeled. In the model, a network grows from an initial configuration by repeatedly choosing a random gene to duplicate. The likelihood that the regulatory relationships associated with the selected gene are retained through duplication is determined by a vector of probabilities. Occurrences of gene-family-specific substructures are counted under the gene duplication model. In this thesis, gene-family-specific substructures are referred to as subnetwork motifs. These subnetwork motifs are motivated by network motifs which are patterns of interconnections that recur more often in a specialized network than in a random network (Milo et al., 2002). Subnetwork motifs differ from network motifs in the way that subnetwork motifs are instances of gene-family-specific substructures while network motifs are isomorphic substructures. These subnetwork motifs are counted under Full and Partial Duplication, which differ in the way in which regulation relationships are inherited. Full duplication occurs when all regulatory links are inherited at each duplication step, and Partial Duplication occurs when regulation inheritance varies at each duplication step. Moments for the number of occurrences of subnetwork motifs are determined in each model. The results presented offer a method for discovering subnetwork motifs that are significant in a GRN under gene duplication.

We report thermal conductivity and Seebeck effect measurements on Mn(Bi1-xSbx)2Te4 (MBST) with x = 0.26 under applied magnetic fields below 50 K. Our data shows clear indications of the electronic structure transition induced by the antiferromagnetic (AFM) to ferromagnetic (FM) transition driven by applied magnetic field as well as significant positive magnetothermal conductivity in the Weyl semimetal state of MBST. Further, by examining the dependence of magnetothermal conductivity on field orientation for MBST and comparison with the magnetothermal conductivity of MnBi2Te4 we see evidence of a contribution to thermal conductivity due to Weyl fermions in the FM phase of MBST. From the temperature dependence of Seebeck coefficient under magnetic fields for MBST, we also observed features consistent with the Fermi surface evolution from a hole pocket in the paramagnetic state to a Fermi surface with coexistence of electron and hole pockets in the FM state. These findings provide further evidence for the field-driven topological phase transition from an AFM topological insulator to a FM Weyl semimetal., Comment: 10 pages, 3 figures

The Wiedemann-Franz (WF) law relates the electronic component of the thermal conductivity to the electrical conductivity in metals through the Lorenz number. The WF law has proven to be remarkably robust, however violations have been observed in many topological materials. In this work, we report thermoelectric measurements conducted on Heusler, ferromagnetic, Weyl semimetal Co$_2$MnAl which shows a drastic, temperature dependent violation of the WF law below 300 K. We then discuss our result in the context of known physical explanations for WF law violation. Both the magnitude and temperature dependence of the violation in Co2MnAl are extreme, indicating that there may be more than one effect contributing to the violation in this system., Comment: 8 pages, 5 figures

We report c-axis transport studies on magnetic topological insulators Mn(Bi1-xSbx)2Te4. We performed systematic c-axis magnetoresistivity measurements under high magnetic fields (up to 35 T) on several representative samples. We find the lightly hole- and lightly electron-doped samples, while both having the same order of magnitude of carrier density and similar spin-flop transitions, exhibit sharp contrast in electronic anisotropy and transport mechanism. The electronic anisotropy is remarkably enhanced for the lightly hole-doped sample relative to pristine MnBi2Te4 but not for the lightly electron-doped sample. The lightly electron-doped sample displays a giant negative longitudinal magnetoresistivity (LMR) induced by the spin-valve effect at the spin-flop transition field, whereas the lightly hole-doped sample exhibits remarkable negative LMR consistent with the chiral anomaly behavior of a Weyl semimetal. Furthermore, we find the large negative LMR of the lightly hole-doped sample extends to a wide temperature range above the N\'eel temperature (T_N) where the magnetoconductivity is proportional to B^2. This fact, together with the short-range intralayer ferromagnetic correlation revealed in isothermal magnetization measurements, suggests the possible presence of the Weyl state above T_N. These results demonstrate that in the c-axis magnetotransport of Mn(Bi1-xSbx)2Te4, the spin scattering is dominant in the lightly electron-doped sample but overwhelmed by the chiral anomaly effect in the lightly hole-doped sample due to the presence of the Weyl state. These findings extend the understanding of the transport properties of Mn(Bi1-xSbx)2Te4., Comment: 21 Pages, 5 Figures

Molt is an essential life-history event in birds and many mammals, as maintenance of feathers and fur is critical for survival. Despite this molt remains an understudied life-history event. Non-standard statistical techniques are required to estimate the phenology of molt from observations of plumage or pelage state, and existing molt phenology models have strict sampling requirements which can be difficult to meet under real-world conditions. We present an extended modelling framework that can accommodate features of real-world molt datasets, such re-encounters of individuals, misclassified molt states, and/or molt state-dependent sampling bias. We demonstrate that such features can lead to biased inferences when using existing molt phenology models, and show that our model extensions can improve inferences about molt phenology under a wide range of sampling conditions. We hope that our novel modelling framework removes barriers for modelling molt phenology data from realworld datasets and thereby further facilitates the uptake of appropriate statistical methods for such data. Although we focus on molt, the modelling framework is applicable to other phenological processes which can be recorded using either ordered categories or approximately linear progress scores., Comment: 44 pages, 15 figures

NLP systems use language models such as Masked Language Models (MLMs) that are pre-trained on large quantities of text such as Wikipedia create representations of language. BERT is a powerful and flexible general-purpose MLM system developed using unlabeled text. Pre-training on large quantities of text also has the potential to transparently embed the cultural and social biases found in the source text into the MLM system. This study aims to compare biases in general purpose and medical MLMs with the StereoSet bias assessment tool. The general purpose MLMs showed significant bias overall, with BERT scoring 57 and RoBERTa scoring 61. The category of gender bias is where the best performances were found, with 63 for BERT and 73 for RoBERTa. Performances for profession, race, and religion were similar to the overall bias scores for the general-purpose MLMs.Medical MLMs showed more bias in all categories than the general-purpose MLMs except for SciBERT, which showed a race bias score of 55, which was superior to the race bias score of 53 for BERT. More gender (Medical 54-58 vs. General 63-73) and religious (46-54 vs. 58) biases were found with medical MLMs. This evaluation of four medical MLMs for stereotyped assessments about race, gender, religion, and profession showed inferior performance to general-purpose MLMs. These medically focused MLMs differ considerably in training source data, which is likely the root cause of the differences in ratings for stereotyped biases from the StereoSet tool., Comment: 5 pages, 1 table

Crystal-field excitations, for example in transition-metal oxides where a rare-earth element is used as a spacer between the transition-metal-oxide tetrahedra and octahedra, are assumed to be extremely robust with respect to external perturbations such as temperature. Using inelastic neutron scattering experiments, a giant shift of the energy of the lowest crystal-field excitation of Er3+ (4I15/2) in ErFeO3 from 0.30(2) meV to 0.75(2) meV was measured below the magnetic-ordering temperature of erbium at 4.1 K. Quantum-mechanical point-charge calculations of the crystal-field levels indicate that the shift is caused by the internal magnetic field created by the erbium spins themselves, which causes a Zeeman splitting of the erbium 4f electronic levels, and therefore a change in the energies of crystal-field transitions. To verify this explanation, the effect of an external magnetic field on the crystal-field excitations was measured by inelastic neutron scattering and compared to the field-dependent point-charge calculations. The existence of an internal magnetic exchange interaction will have implications for a deeper understanding of a broader group of phenomena such as multiferroic properties or spin frustration, which are a consequence of various competing electronic and magnetic exchange interactions., Comment: Before submission we had contacted all co-authors repeatedly if they have any scientific comments, questions, or concerns. After submission one co-author did rise scientific concerns. Since we were not able to solve this, we would like to withdraw our manuscript

Advances in space weather science and small satellite (SmallSat) technology have proceeded in parallel over the past two decades, but better communication and coordination is needed among the respective worldwide communities contributing to this rapid progress. We identify six areas where improved international coordination is especially desirable, including: (1) orbital debris mitigation; (2) spectrum management; (3) export control regulations; (4) access to timely and low-cost launch opportunities; (5) inclusive data policies; and (6) education. We argue the need for internationally coordinated policies and programs to promote the use of SmallSats for space weather research and forecasting while realizing maximum scientific and technical advances through the integration of these two increasingly important endeavors., Comment: Commentary submitted to Space Weather; accepted and in press

This is an empirical study to investigate the impact of scanner effects when using machine learning on multi-site neuroimaging data. We utilize structural T1-weighted brain MRI obtained from two different studies, Cam-CAN and UK Biobank. For the purpose of our investigation, we construct a dataset consisting of brain scans from 592 age- and sex-matched individuals, 296 subjects from each original study. Our results demonstrate that even after careful pre-processing with state-of-the-art neuroimaging pipelines a classifier can easily distinguish between the origin of the data with very high accuracy. Our analysis on the example application of sex classification suggests that current approaches to harmonize data are unable to remove scanner-specific bias leading to overly optimistic performance estimates and poor generalization. We conclude that multi-site data harmonization remains an open challenge and particular care needs to be taken when using such data with advanced machine learning methods for predictive modelling., Comment: Presented at the Medical Imaging meets NeurIPS Workshop 2019

Background: The trend towards large-scale studies including population imaging poses new challenges in terms of quality control (QC). This is a particular issue when automatic processing tools, e.g. image segmentation methods, are employed to derive quantitative measures or biomarkers for later analyses. Manual inspection and visual QC of each segmentation isn't feasible at large scale. However, it's important to be able to automatically detect when a segmentation method fails so as to avoid inclusion of wrong measurements into subsequent analyses which could lead to incorrect conclusions. Methods: To overcome this challenge, we explore an approach for predicting segmentation quality based on Reverse Classification Accuracy, which enables us to discriminate between successful and failed segmentations on a per-cases basis. We validate this approach on a new, large-scale manually-annotated set of 4,800 cardiac magnetic resonance scans. We then apply our method to a large cohort of 7,250 cardiac MRI on which we have performed manual QC. Results: We report results used for predicting segmentation quality metrics including Dice Similarity Coefficient (DSC) and surface-distance measures. As initial validation, we present data for 400 scans demonstrating 99% accuracy for classifying low and high quality segmentations using predicted DSC scores. As further validation we show high correlation between real and predicted scores and 95% classification accuracy on 4,800 scans for which manual segmentations were available. We mimic real-world application of the method on 7,250 cardiac MRI where we show good agreement between predicted quality metrics and manual visual QC scores. Conclusions: We show that RCA has the potential for accurate and fully automatic segmentation QC on a per-case basis in the context of large-scale population imaging as in the UK Biobank Imaging Study., Comment: 14 pages, 7 figures, Journal of Cardiovascular Magnetic Resonance

Recent advances in deep learning based image segmentation methods have enabled real-time performance with human-level accuracy. However, occasionally even the best method fails due to low image quality, artifacts or unexpected behaviour of black box algorithms. Being able to predict segmentation quality in the absence of ground truth is of paramount importance in clinical practice, but also in large-scale studies to avoid the inclusion of invalid data in subsequent analysis. In this work, we propose two approaches of real-time automated quality control for cardiovascular MR segmentations using deep learning. First, we train a neural network on 12,880 samples to predict Dice Similarity Coefficients (DSC) on a per-case basis. We report a mean average error (MAE) of 0.03 on 1,610 test samples and 97% binary classification accuracy for separating low and high quality segmentations. Secondly, in the scenario where no manually annotated data is available, we train a network to predict DSC scores from estimated quality obtained via a reverse testing strategy. We report an MAE=0.14 and 91% binary classification accuracy for this case. Predictions are obtained in real-time which, when combined with real-time segmentation methods, enables instant feedback on whether an acquired scan is analysable while the patient is still in the scanner. This further enables new applications of optimising image acquisition towards best possible analysis results., Comment: Accepted at MICCAI 2018

Web tracking technologies are pervasive and operated by a few large technology companies. This technology, and the use of the collected data has been implicated in influencing elections, fake news, discrimination, and even health decisions. Little is known about how this technology is deployed on hospital or other health related websites. The websites of the 210 public hospitals in the state of Illinois, USA were evaluated with a web tracker identification tool. Web trackers were identified on 94% of hospital webs sites, with an average of 3.5 trackers on the websites of general hospitals. The websites of smaller critical access hospitals used an average of 2 web trackers. The most common web tracker identified was Google Analytics, found on 74% of Illinois hospital websites. Of the web trackers discovered, 88% were operated by Google and 26% by Facebook. In light of revelations about how web browsing profiles have been used and misused, search bubbles, and the potential for algorithmic discrimination hospital leadership and policy makers must carefully consider if it is appropriate to use third party tracking technology on hospital web sites., Comment: 7 pages, 1 table, 2 figures

The Hypertext Transfer Protocol Secure (HTTPS) communications protocol is used to secure traffic between a web browser and server. This technology can significantly reduce the risk of interception and manipulation of web information for nefarious purposes such as identity theft. Deployment of HTTPS has reached about 50% of all webs sites. Little is known about HTTPS implantation for hospital websites. To investigate the prevalence of HTTPS implementation, we analyzed the websites of the 210 public hospitals in the state of Illinois, USA. HTTPS was implemented to industry standards for 54% of all hospital websites in Illinois. Geographical analysis showed an urban vs. rural digital divide with 60% of urban hospitals and 40% of rural hospitals implementing HTTPS., Comment: 5 pages, 1 table. arXiv admin note: text overlap with arXiv:1712.05376

The domain name system translates human friendly web addresses to a computer readable internet protocol address. This basic infrastructure is insecure and can be manipulated. Deployment of technology to secure the DNS system has been slow, reaching about 20% of all web sites based in the USA. Little is known about the efforts hospitals and health systems make to secure the domain name system for their websites. To investigate the prevalence of implementing Domain Name System Security Extensions (DNSSEC), we analyzed the websites of the 210 public hospitals in the state of Illinois, USA. Only one Illinois hospital website was found to have implemented DNSSEC by December, 2017., Comment: 4 pages