The inflammable growth of misinformation on social media and other platforms during pandemic situations like COVID-19 can cause significant damage to the physical and mental stability of the people. To detect such misinformation, researchers have been applying various machine learning (ML) and deep learning (DL) techniques. The objective of this study is to systematically review, assess, and synthesize state-of-the-art research articles that have used different ML and DL techniques to detect COVID-19 misinformation. A structured literature search was conducted in the relevant bibliographic databases to ensure that the survey solely centered on reproducible and high-quality research. We reviewed 43 papers that fulfilled our inclusion criteria out of 260 articles found from our keyword search. We have surveyed a complete pipeline of COVID-19 misinformation detection. In particular, we identify various COVID-19 misinformation datasets and review different data processing, feature extraction, and classification techniques to detect COVID-19 misinformation. At the end, the challenges and limitations in detecting COVID-19 misinformation using machine learning techniques and the future research directions are discussed.
The patient population suffering from pancreatic ductal adenocarcinoma (PDAC) presents, as a whole, with a high degree of molecular tumor heterogeneity. The heterogeneity of PDAC tumor composition has complicated treatment and stalled success in clinical trials. Current in vitro techniques insufficiently replicate the intricate stromal components of PDAC tumor microenvironments (TMEs) and fail to model a given tumor’s unique genetic phenotype. The development of patient-derived organoids (PDOs) has opened the door for improved personalized medicine since PDOs are derived directly from patient tumors, thus preserving the tumors’ unique behaviors and genetic phenotypes. This study developed a tumor-chip device engineered to mimic the PDAC TME by incorporating PDOs and stromal cells, specifically pancreatic stellate cells and macrophages. Establishing PDOs in a multicellular microfluidic chip device prolongs cellular function and longevity and successfully establishes a complex organotypic tumor environment that incorporates desmoplastic stroma and immune cells. When primary cancer cells in monoculture were subjected to stroma-depleting agents, there was no effect on cancer cell viability. However, targeting stroma in our tumor-chip model resulted in a significant increase in the chemotherapy effect on cancer cells, thus validating the use of this tumor-chip device for drug testing.
Background Theoretically, artificial intelligence can provide an accurate automatic solution to measure right ventricular (RV) ejection fraction (RVEF) from cardiovascular magnetic resonance (CMR) images, despite the complex RV geometry. However, in our recent study, commercially available deep learning (DL) algorithms for RVEF quantification performed poorly in some patients. The current study was designed to test the hypothesis that quantification of RV function could be improved in these patients by using more diverse CMR datasets in addition to domain-specific quantitative performance evaluation metrics during the cross-validation phase of DL algorithm development. Methods We identified 100 patients from our prior study who had the largest differences between manually measured and automated RVEF values. Automated RVEF measurements were performed using the original version of the algorithm (DL1), an updated version (DL2) developed from a dataset that included a wider range of RV pathology and validated using multiple domain-specific quantitative performance evaluation metrics, and conventional methodology performed by a core laboratory (CORE). Each of the DL-RVEF approaches was compared against CORE-RVEF reference values using linear regression and Bland–Altman analyses. Additionally, RVEF values were classified into 3 categories: ≤ 35%, 35–50%, and ≥ 50%. Agreement between RVEF classifications made by the DL approaches and the CORE measurements was tested. Results CORE-RVEF and DL-RVEFs were obtained in all patients (feasibility of 100%). DL2-RVEF correlated with CORE-RVEF better than DL1-RVEF (r = 0.87 vs. r = 0.42), with narrower limits of agreement. As a result, DL2 algorithm also showed increasing accuracy from 0.53 to 0.80 for categorizing RV function. Conclusions The use of a new DL algorithm cross-validated on a dataset with a wide range of RV pathology using multiple domain-specific metrics resulted in a considerable improvement in the accuracy of automated RVEF measurements. This improvement was demonstrated in patients whose images were the most challenging and resulted in the largest RVEF errors. These findings underscore the critical importance of this strategy in the development of DL approaches for automated CMR measurements.
High entropy alloys (HEAs) are an important material class in the development of next-generation structural materials, but the astronomically large composition space cannot be efficiently explored by experiments or first-principles calculations. Machine learning (ML) methods might address this challenge, but ML of HEAs has been hindered by the scarcity of HEA property data. In this work, the EMTO-CPA method was used to generate a large HEA dataset (spanning a composition space of 14 elements) containing 7086 cubic HEA structures with structural properties, 1911 of which have the complete elastic tensor calculated. The elastic property dataset was used to train a ML model with the Deep Sets architecture. The Deep Sets model has better predictive performance and generalizability compared to other ML models. Association rule mining was applied to the model predictions to describe the compositional dependence of HEA elastic properties and to demonstrate the potential for data-driven alloy design.
Bismuth-based materials (e.g., metallic, oxides and subcarbonate) are emerged as promising electrocatalysts for converting CO 2 to formate. However, Bi o -based electrocatalysts possess high overpotentials, while bismuth oxides and subcarbonate encounter stability issues. This work is designated to exemplify that the operando synthesis can be an effective means to enhance the stability of electrocatalysts under operando CO 2 RR conditions. A synthetic approach is developed to electrochemically convert BiOCl into Cl-containing subcarbonate (Bi 2 O 2 (CO 3 ) x Cl y ) under operando CO 2 RR conditions. The systematic operando spectroscopic studies depict that BiOCl is converted to Bi 2 O 2 (CO 3 ) x Cl y via a cathodic potential-promoted anion-exchange process. The operando synthesized Bi 2 O 2 (CO 3 ) x Cl y can tolerate − 1.0 V versus RHE, while for the wet-chemistry synthesized pure Bi 2 O 2 CO 3 , the formation of metallic Bi o occurs at − 0.6 V versus RHE. At − 0.8 V versus RHE, Bi 2 O 2 (CO 3 ) x Cl y can readily attain a FE HCOO - of 97.9%, much higher than that of the pure Bi 2 O 2 CO 3 (81.3%). DFT calculations indicate that differing from the pure Bi 2 O 2 CO 3 -catalyzed CO 2 RR, where formate is formed via a * OCHO intermediate step that requires a high energy input energy of 2.69 eV to proceed, the formation of HCOO ⁻ over Bi 2 O 2 (CO 3 ) x Cl y has proceeded via a * COOH intermediate step that only requires low energy input of 2.56 eV.
SAXS models, used to calculate the micro-textural properties (porosity, specific surface, correlation length, correlation function) of porous materials, are especially relevant for in-situ material characterization. Certain classes of porous materials, including those generated during solid decompositions, are characterized by distinct regions with different micro-textures. Classical SAXS models cannot be applied directly to such multi-texture systems, thus a novel SAXS model for multi-texture systems is proposed in this work and validated by using measurements of micro-textural properties of a calcined CaCO3 powder, obtained using time-resolved in-situ synchrotron radiation USAXS, SAXS and WAXS data. The proposed model, in contrast to the direct application of classical SAXS models, is capable of correctly predicting the expected linear trend of the particle internal porosity with respect to the CaCO3 conversion and provides a complete description of the micro-textural properties, in very close agreement with experimental data available in the literature. In the reaction product region the correlation function, as well as the other micro-textural properties, is essentially constant during the solid decomposition, and is accurately represented by a two-parameter model. Based on the proposed SAXS model, a simplified expression of the differential scattering cross section, as a product of the CaO mass fraction times a time-independent function of the correlation function, is obtained; such expression can reliably predict the experimental scattering intensity profiles over time.
Distribution network operation is becoming more challenging because of the growing integration of intermittent and volatile distributed energy resources (DERs). This motivates the development of new distribution system state estimation (DSSE) paradigms that can operate at fast timescale based on real-time data stream of asynchronous measurements enabled by modern information and communications technology. To solve the real-time DSSE with asynchronous measurements effectively and accurately, this paper formulates a weighted least squares DSSE problem and proposes an online stochastic gradient algorithm to solve it. The performance of the proposed scheme is analytically guaranteed and is numerically corroborated with realistic data on IEEE 123-bus feeder.
We present a Compound Poisson Mixture Regression model of the joint distribution of transaction frequency and monetary value, and apply it to study alumni donations at a university in the USA. The model captures covariate effects, recognizing that both response variables emanate from one statistical unit — a donor. Heterogeneity, group-level factors, and other features of the data are captured through coefficients that vary between segments. The data in the study are transaction records for the 2000–2016 period, and a survey conducted in 2017. Despite including subjective factors from the survey, the results suggest that between-segment differences are unobserved. Heterogeneity is manifested in covariates, including subjective factors – psychological distance, perceptions of donation impact, willingness to volunteer – displaying stratified effects on either transaction amounts, frequencies, or compound effects on both variables. Characterization of such effects supports the development of tailored fundraising/marketing strategies aimed at increasing donor retention and lifetime value.
DP-coloring (also called correspondence coloring) is a generalization of list coloring that has been widely studied in recent years after its introduction by Dvořák and Postle in 2015. As the analogue of the chromatic polynomial P(G,m), the DP color function of a graph G, denoted PDP(G,m), counts the minimum number of DP-colorings over all possible m-fold covers. Chromatic polynomials for joins and vertex-gluings of graphs are well understood, but the effect of these graph operations on the DP color function is not known. In this paper we make progress on understanding the DP color function of the join of a graph with a complete graph and vertex-gluings of certain graphs. We also develop tools to study the DP color function under these graph operations, and we study the threshold (smallest m) beyond which the DP color function of a graph constructed with these operations equals its chromatic polynomial.
International Electrotechnical Commission (IEC) proposed the IEC three-ratio method based on Dissolved Gas Analysis (DGA), which is one of the most effective tools for Power Transformer Fault Diagnosis (PTFD). However, the PTFD accuracy is generally limited because the classification boundary could be too stiff to classify samples located near the boundary. The Support Vector Machine (SVM) was applied to PTFD to improve diagnosis accuracy, while traditional SVM multi-classification methods and parameter optimization algorithms are subject to poor training efficiency. As a result, the SVM-based PTFD model is difficult to update frequently with the accumulation of fault data. A new SVM-based PTFD decision framework is proposed in this paper which can significantly boost the training efficiency and ensure the accuracy. In the proposed framework, a multi-step feature extraction process consisting of characteristic gas concentration and its ratios is applied. Based on the feature distribution of various samples, a proper SVM multi-classification method is presented using a hierarchical decision tree structure. In addition, according to the principles of SVM and radial basis kernel function, a Support Vector feature-based parameter optimization algorithm (SVFB) is proposed. IEC TC 10 data and the historical data of online transformer monitoring provided by the State Grid Corporation of China are adopted as sample sets. The simulation results demonstrate that the proposed decision framework can reach high diagnosis accuracy while shortening the training time.
Why is it that both complex and simple solutions that have proved to be effective have low rates of adoption? The literature on innovation (i.e., a specific category of solutions) management has provided some clues, identifying barriers of several types: organizational, technological, economic, human behavior and the nature of the innovation. We suggest that one reason is the misalignment between the degrees of complexity i.e., the degree of knowledge embedded, of the problem and its solution. A solution perceived to be too simple for a complex problem falls into the category of what might be called “Columbus' egg”. At the basis of this effect there is the tendency to minimize expected frustration as the difference between the effort made in looking for a solution and the obtained reward. When the solution is too complex for a simple problem, this is the case of the “Engineer's effect”. This effect has its cognitive underpinnings in the tendency to minimize decision-making costs. We discuss and illustrate these phenomena and propose some guidelines for technology developers and product innovation managers, as well as for forecasting solutions adoption.
The impact of legislation in shaping social norms has captured both scholarly and practitioner attention in the past decades. However, limited understanding exists on how social legislation can create economic value for firms and thereby strengthen the business case for such legislation. We attempt to theorize and test this phenomenon in the context of marriage equality. We first attempt to understand the role of state-wise same-sex marriage social legislation on financial performance of firms headquartered in states where the legislation is passed. Adapting the model of institutional racism to the context of heterosexism, we then develop a framework to examine the role of societal-cultural, institutional, and individual factors in shaping the effect of state-wise same-sex marriage legislation on firm performance. In a sample of publicly traded U.S. firms, we conduct a difference-in-difference analysis to test and find support for our arguments.
Background Pre-operative patient education has been shown to reduce patient anxiety and increase patient medical comprehension. This prospective pilot study hypothesized that patient knowledge related to the radiotherapy treatment process would be low after receiving traditional radiation educational materials and counseling at the time of initial consultation. Methods Patients with non-metastatic cancer receiving definitive or adjuvant radiotherapy at three suburban radiation therapy clinics affiliated with an academic medical center completed a 34-question survey. Patients received traditional radiation educational materials. The survey included questions on demographics, the Spielberger State-Trait Anxiety Inventory short-form (STAI-S-6), a modified radiotherapy Amsterdam preoperative anxiety and information scale (mRT-APAIS), and radiotherapy knowledge. Patients also provided qualitative responses. Surveys were administered prior to the patient's CT-simulation scan. Descriptive statistics were performed. Results 22 patients were enrolled in this prospective pilot study. 19 (86%) patients were female. The median age was 66 (range: 38-80). 14 (64%) patients were white, 7 (32%) black, and 1 (4%) American Indian/Alaskan Native. 6 (27%) patients received a high school degree or GED, 10 (46%) obtained a 2-year degree and 5 (22%) received a 4-year degree or higher. 16 (73%) patients had breast cancer with the others having lung, brain, gastrointestinal, gynecologic, or other malignancies. The median radiotherapy knowledge score was 35.0 [IQR: 31.0 - 39.5] and 58% of patients endorsed low overall radiotherapy knowledge (score < 35). Patients reported low levels of knowledge related to radiotherapy set-up (41%), immobilization (41%), x-ray use (64%), and sensation (tactile 41%, auditory: 59%). The median STAI-S-6 score was 43.3 [IQR: 36.7 - 46.7] and the median mRT-APAIS score was 18 [IQR: 14.8 - 20.0]. 68% of patients were "anxious" by STAI-S-6 (score ≥ 40) and 77% by mRT-APAIS (score ≥ 12). Discussion A majority of patients reported low levels of knowledge related to the radiotherapy treatment process prior to CT simulation. These findings suggest an educational intervention at the time of consultation may be beneficial to reduce patient anxiety and increase patient knowledge related to the radiotherapy treatment process. Accrual is ongoing in this prospective pilot study. Future research directions include a phase 3 multi-institutional stepped-wedge clinical trial to investigate the impact of a novel patient education tool such as the Communicating the External Beam Radiotherapy Experience (CEBRE) discussion guide used during initial consultation to reduce patient anxiety and increase patient knowledge about radiotherapy.
- Behzad Imanian
- John Donaghy
- Tim Jackson
- Jason Wan
The development and application of modern sequencing technologies have led to many new improvements in food safety and public health. With unprecedented resolution and big data, high-throughput sequencing (HTS) has enabled food safety specialists to sequence marker genes, whole genomes, and transcriptomes of microorganisms almost in real-time. These data reveal not only the identity of a pathogen or an organism of interest in the food supply but its virulence potential and functional characteristics. HTS of amplicons, allow better characterization of the microbial communities associated with food and the environment. New and powerful bioinformatics tools, algorithms, and machine learning allow for development of new models to predict and tackle important events such as foodborne disease outbreaks. Despite its potential, the integration of HTS into current food safety systems is far from complete. Government agencies have embraced this new technology, and use it for disease diagnostics, food safety inspections, and outbreak investigations. However, adoption and application of HTS by the food industry have been comparatively slow, sporadic, and fragmented. Incorporation of HTS by food manufacturers in their food safety programs could reinforce the design and verification of effectiveness of control measures by providing greater insight into the characteristics, origin, relatedness, and evolution of microorganisms in our foods and environment. Here, we discuss this new technology, its power, and potential. A brief history of implementation by public health agencies is presented, as are the benefits and challenges for the food industry, and its future in the context of food safety.
- Saran Pidaparthy
- Mei Luo
- Marco-Tulio F. Rodrigues
- Daniel P. Abraham
The severe capacity fade of lithium-ion cells with silicon-dominant anodes has hindered their widescale commercialization. In this work, we link cell capacity fade to the heterogeneous physicochemical evolution of silicon anodes during battery cycling. Through a multilength scale characterization approach, we demonstrate that silicon particles near the anode surface react differently from those near the copper current collector. In particular, near the anode surface we find an amorphized wispy silicon encased in a highly fluorinated matrix of electrolyte-reduction products. In contrast, closer to the current collector, the silicon retains more of its initial morphology and structure, suggesting the presence of isolated particles. The results show that the accessibility of active silicon to lithium ions varies across the anode matrix. Material and cell designs, which minimize electrode expansion resulting from the in-filling of pores with the solid electrolyte interphase (SEI), are needed to enhance anode homogeneity during the electrochemical cycling.
Bax is a well-known universal proapoptotic protein. Bax protein is detected in almost all human organs, and its expression levels can be correlated with disease progression and therapeutic efficacy in certain settings. Interestingly, increasing evidence has shown that mature neuronal cell death is often not typical apoptosis. Most results on the expression of Bax proteins (predominantly Baxα) in the human brain come from disease-oriented studies, and the data on Bax protein expression in the normal brain are limited and lack consistency due to many variable factors. Here, we analyzed Bax RNA and protein expression data from multiple databases and performed immunostaining of over 80 samples from 25 healthy subjects across 7 different brain regions. We found that Bax protein expression was heterogeneous across brain regions and individual subjects. Both neurons and glial cells, such as astrocytes, could be Bax positive, but Bax positivity appeared to be highly selective, even within the same cell type in the same region. Furthermore, Bax proteins could be localized in the cytosol (evenly spread or concentrated to one region), nucleus or nucleolus depending on the cell type. Such variation and distribution in Bax expression suggest that Bax may function differently in the human brain than in other organs.
With the rapid development of computational techniques and scientific tools, great progress of data-driven analysis has been made to extract governing laws of dynamical systems from data. Despite the wide occurrences of non-Gaussian fluctuations, the effective data-driven methods to identify stochastic differential equations with non-Gaussian Lévy noise are relatively few so far. In this work, we propose a data-driven approach to extract stochastic governing laws with both (Gaussian) Brownian motion and (non-Gaussian) Lévy motion, from short bursts of simulation data. Specifically, we use the normalizing flows technology to estimate the transition probability density function (solution of non-local Fokker–Planck equations) from data, and then substitute it into the recently proposed non-local Kramers–Moyal formulae to approximate Lévy jump measure, drift coefficient and diffusion coefficient. We demonstrate that this approach can learn the stochastic differential equation with Lévy motion. We present examples with one- and two-dimensional decoupled and coupled systems to illustrate our method. This approach will become an effective tool for discovering stochastic governing laws and understanding complex dynamical behaviours. This article is part of the theme issue ‘Data-driven prediction in dynamical systems’.
Successful monitoring of the condition of stimulation electrodes is critical for maintaining chronic device performance for neural stimulation. As part of pre-clinical safety testing in preparation for a visual prostheses clinical trial, we evaluated the stability of the implantable devices and stimulation electrodes using a combination of current pulsing in saline and in canine visual cortex. Specifically, in saline we monitored the stability and performance of 3000 μm ² geometric surface area activated iridium oxide film (AIROF) electrodes within a wireless floating microelectrode array (WFMA) by measuring the voltage transient (VT) response through reverse telemetry. Eight WFMAs were assessed in vitro for 24 days, where n = 4 were pulsed continuously at 80 μA (16 nC/phase) and n = 4 remained in solution with no applied stimulation. Subsequently, twelve different WFMAs were implanted in visual cortex in n = 3 canine subjects (4 WFMAs each). After a 2-week recovery period, half of the electrodes in each of the twelve devices were pulsed continuously for 24 h at either 20, 40, 63, or 80 μA (200 μs pulse width, 100 Hz). VTs were recorded to track changes in the electrodes at set time intervals in both the saline and in vivo study. The VT response of AIROF electrodes remained stable during pulsing in saline over 24 days. Electrode polarization and driving voltage changed by less than 200 mV on average. The AIROF electrodes also maintained consistent performance, overall, during 24 h of pulsing in vivo . Four of the in vivo WFMA devices showed a change in polarization, access voltage, or driving voltage over time. However, no VT recordings indicated electrode failure, and the same trend was typically seen in both pulsed and unpulsed electrodes within the same device. Overall, accelerated stimulation testing in saline and in vivo indicated that AIROF electrodes in the WFMA were able to consistently deliver up to 16 nC per pulse and would be suitable for chronic clinical use.
Institution pages aggregate content on ResearchGate related to an institution. The members listed on this page have self-identified as being affiliated with this institution. Publications listed on this page were identified by our algorithms as relating to this institution. This page was not created or approved by the institution. If you represent an institution and have questions about these pages or wish to report inaccurate content, you can contact us here.