Scientific Knowledge Discovery Research Articles

LitAI: Enhancing Multimodal Literature Understanding and Mining with Generative AI.

Information processing and retrieval in literature are critical for advancing scientific research and knowledge discovery. The inherent multimodality and diverse literature formats, including text, tables, and figures, present significant challenges in literature information retrieval. This paper introduces LitAI, a novel approach that employs readily available generative AI tools to enhance multimodal information retrieval from literature documents. By integrating tools such as optical character recognition (OCR) with generative AI services, LitAI facilitates the retrieval of text, tables, and figures from PDF documents. We have developed specific prompts that leverage in-context learning and prompt engineering within Generative AI to achieve precise information extraction. Our empirical evaluations, conducted on datasets from the ecological and biological sciences, demonstrate the superiority of our approach over several established baselines including Tesseract-OCR and GPT-4. The implementation of LitAI is accessible at https://github.com/ResponsibleAILab/LitAI.

Current and future applications of artificial intelligence in surgery: implications for clinical practice and research.

Surgeons are skilled at making complex decisions over invasive procedures that can save lives and alleviate pain and avoid complications in patients. The knowledge to make these decisions is accumulated over years of schooling and practice. Their experience is in turn shared with others, also via peer-reviewed articles, which get published in larger and larger amounts every year. In this work, we review the literature related to the use of Artificial Intelligence (AI) in surgery. We focus on what is currently available and what is likely to come in the near future in both clinical care and research. We show that AI has the potential to be a key tool to elevate the effectiveness of training and decision-making in surgery and the discovery of relevant and valid scientific knowledge in the surgical domain. We also address concerns about AI technology, including the inability for users to interpret algorithms as well as incorrect predictions. A better understanding of AI will allow surgeons to use new tools wisely for the benefit of their patients.

Knowledge graph-guided data-driven design of ultra-high-performance concrete (UHPC) with interpretability and physicochemical reaction discovery capability

Traditional methods for designing concrete materials typically rely on labor-intensive laboratory experiments, resulting in time and cost inefficiencies. Recently, designing concrete using artificial intelligence (AI) methods has shown high efficiency, but existing AI methods often rely solely on data, which can lead to violation with scientific principles and result in models lacking reasoning abilities. To overcome these challenges, this paper presents an interpretable knowledge graph-guided data-driven design approach. By integrating advanced computing techniques with domain knowledge via knowledge graphs, this approach enables the interpretation of data-driven models and uncovers the underlying mechanisms behind predictions. This approach is applied to ultra-high-performance concrete (UHPC) involving complex physicochemical reactions. The domain knowledge about UHPC is imparted using a knowledge graph, and UHPC properties are predicted using a machine learning model considering mixing proportions, processing methods, and physiochemical properties of materials via natural language processing. The results show that the knowledge graph displays crucial design variables and their effects on UHPC properties, aiding in selecting variables for machine learning models and interpreting their results. The prediction accuracy of the machine learning model reached 0.95. The research paves the way for more transparent and scientific AI models for material design and AI-enabled discovery of scientific knowledge.

Deep learning discovery of macroscopic governing equations for viscous gravity currents from microscopic simulation data

Although deep learning has been successfully applied in a variety of science and engineering problems owing to its strong high-dimensional nonlinear mapping capability, it is of limited use in scientific knowledge discovery. In this work, we propose a deep learning based framework to discover the macroscopic governing equation of an important geophysical process, i.e., viscous gravity current, based on high-resolution microscopic simulation data without the need for prior knowledge of underlying terms. For two typical scenarios with different viscosity ratios, the deep learning based equations exactly capture the same dominant terms as the theoretically derived equations for describing long-term asymptotic behaviors, which validates the proposed framework. Unknown macroscopic equations are then obtained for describing short-term behaviors, and additional deep-learned compensation terms are eventually discovered. Comparison of posterior tests shows that the deep learning based PDEs actually perform better than the theoretically derived PDEs in predicting evolving viscous gravity currents for both long-term and short-term regimes. Moreover, the proposed framework is proven to be very robust against non-biased data noise for training, which is up to 20%. Consequently, the presented deep learning framework shows considerable potential for discovering unrevealed intrinsic laws in scientific semantic space from raw experimental or simulation results in data space.

Mapping scientific knowledge discovery on COVID-19 pandemic and agriculture: a bibliometric analysis and future research directions.

This paper aims at analyzing the research productivity and scientific knowledge discovery of the COVID-19 pandemic in agriculture using a bibliometric analysis approach. A total of 1514 research papers indexed in the Scopus database, covering a period of 2020 to 2022, are processed using VOSviewer and R-Studio software. The analysis of research productivity indicates that the number of research publications on COVID-19 and agriculture has exponentially increased globally, and about 80% of the research papers have been published in the top 10 countries led by the USA, India, and China. The countries are increasingly collaborating in undertaking research on COVID-19 and agriculture. Furthermore, major journals and articles with citations have been extracted to analyze the leading publication avenues and focused areas of research. The science mapping is done using co-occurrence and thematic map. With the help of co-occurrence analysis, six clusters are identified depicting major research themes, i.e., COVID-19 and agricultural supply chain disruption, COVID-19 and human health issues and coping strategies, COVID-19 and non-human and animal health, COVID-19 pandemic and environment and pollution, COVID-19 and healthcare and treatment, and COVID-19 and food nutrition from dairy and meat products. The thematic map analysis identifies potential research areas such as mental health, anxiety, and depression in the agricultural system, which may help in setting future research agenda and help devising policy supports for managing the agriculture sector better during crisis. The paper also highlights the theoretical and practical implications.

Predicting 30-Day Readmission Risk for Patients With Chronic Obstructive Pulmonary Disease Through a Federated Machine Learning Architecture on Findable, Accessible, Interoperable, and Reusable (FAIR) Data: Development and Validation Study.

BackgroundOwing to the nature of health data, their sharing and reuse for research are limited by legal, technical, and ethical implications. In this sense, to address that challenge and facilitate and promote the discovery of scientific knowledge, the Findable, Accessible, Interoperable, and Reusable (FAIR) principles help organizations to share research data in a secure, appropriate, and useful way for other researchers.ObjectiveThe objective of this study was the FAIRification of existing health research data sets and applying a federated machine learning architecture on top of the FAIRified data sets of different health research performing organizations. The entire FAIR4Health solution was validated through the assessment of a federated model for real-time prediction of 30-day readmission risk in patients with chronic obstructive pulmonary disease (COPD).MethodsThe application of the FAIR principles on health research data sets in 3 different health care settings enabled a retrospective multicenter study for the development of specific federated machine learning models for the early prediction of 30-day readmission risk in patients with COPD. This predictive model was generated upon the FAIR4Health platform. Finally, an observational prospective study with 30 days follow-up was conducted in 2 health care centers from different countries. The same inclusion and exclusion criteria were used in both retrospective and prospective studies.ResultsClinical validation was demonstrated through the implementation of federated machine learning models on top of the FAIRified data sets from different health research performing organizations. The federated model for predicting the 30-day hospital readmission risk was trained using retrospective data from 4.944 patients with COPD. The assessment of the predictive model was performed using the data of 100 recruited (22 from Spain and 78 from Serbia) out of 2070 observed (records viewed) patients during the observational prospective study, which was executed from April 2021 to September 2021. Significant accuracy (0.98) and precision (0.25) of the predictive model generated upon the FAIR4Health platform were observed. Therefore, the generated prediction of 30-day readmission risk was confirmed in 87% (87/100) of cases.ConclusionsImplementing a FAIR data policy in health research performing organizations to facilitate data sharing and reuse is relevant and needed, following the discovery, access, integration, and analysis of health research data. The FAIR4Health project proposes a technological solution in the health domain to facilitate alignment with the FAIR principles.

Physics-guided machine learning: A new paradigm for scientific knowledge discovery

Physics-guided machine learning: A new paradigm for scientific knowledge discovery

Automated Deduction and Knowledge Management in Geometry

Scientific research and education at all levels are concerned primarily with the discovery, verification, communication, and application of scientific knowledge. Learning, reusing, inventing, and archiving are the four essential aspects of knowledge accumulation in mankind’s civilisation process. In this cycle of knowledge accumulation, which has been supported for thousands of years by written books and other physical means, rigorous reasoning has always played an essential role. Nowadays this process is becoming more and more effective due to the availability of new paradigms based on computer applications. Geometric reasoning with such computer applications is one of the most attractive challenges for future accumulation and dissemination of knowledge.

A Scientific Knowledge Discovery and Data Mining Process Model for Metabolomics

This work presents a scientific data mining process model for metabolomics that provides a systematic and formalised framework for guiding and performing metabolomics data analysis in a justifiable and traceable manner. The process model is designed to promote the achievement of the analytical objectives of metabolomics investigations and to ensure the validity, interpretability and reproducibility of their results. It satisfies the requirements of metabolomics data mining, focuses on the contextual meaning of metabolomics knowledge, and addresses the shortcomings of existing data mining process models, while paying attention to the practical aspects of metabolomics investigations and other desirable features. The process model development involved investigating the ontologies and standards of science, data mining and metabolomics and its design was based on the principles, best practices and inspirations from Process Engineering, Software Engineering, Scientific Methodology and Machine Learning. A software environment was built to realise and automate the process model execution and was then applied to a number of metabolomics datasets to demonstrate and evaluate its applicability to different metabolomics investigations, approaches and data acquisition instruments on one hand, and to different data mining approaches, goals, tasks and techniques on the other. The process model was successful in satisfying the requirements of metabolomics data mining and can be generalised to perform data mining in other scientific disciplines.

Scientific discovery, causal explanation, and process model induction

In this paper, I review two related lines of computational research: discovery of scientific knowledge and causal models of scientific phenomena. I also report research on quantitative process models that falls at the intersection of these two themes. This framework represents models as a set of interacting processes, each with associated differential equations that express influences among variables. Simulating such a quantitative process model produces trajectories for variables over time that one can compare to observations. Background knowledge about candidate processes enables search through the space of model structures and associated parameters to find explanations of time-series data. I discuss the representation of such process models, their use for prediction and explanation, and their discovery through heuristic search, along with their interpretation as causal accounts of dynamic behavior.

Mechanistic Computational Models of MicroRNA-Mediated Signaling Networks in Human Diseases.

MicroRNAs (miRs) are endogenous non-coding RNA molecules that play important roles in human health and disease by regulating gene expression and cellular processes. In recent years, with the increasing scientific knowledge and new discovery of miRs and their gene targets, as well as the plentiful experimental evidence that shows dysregulation of miRs in a wide variety of human diseases, the computational modeling approach has emerged as an effective tool to help researchers identify novel functional associations between differential miR expression and diseases, dissect the phenotypic expression patterns of miRs in gene regulatory networks, and elucidate the critical roles of miRs in the modulation of disease pathways from mechanistic and quantitative perspectives. Here we will review the recent systems biology studies that employed different kinetic modeling techniques to provide mechanistic insights relating to the regulatory function and therapeutic potential of miRs in human diseases. Some of the key computational aspects to be discussed in detail in this review include (i) models of miR-mediated network motifs in the regulation of gene expression, (ii) models of miR biogenesis and miR–target interactions, and (iii) the incorporation of such models into complex disease pathways in order to generate mechanistic, molecular- and systems-level understanding of pathophysiology. Other related bioinformatics tools such as computational platforms that predict miR-disease associations will also be discussed, and we will provide perspectives on the challenges and opportunities in the future development and translational application of data-driven systems biology models that involve miRs and their regulatory pathways in human diseases.

Cardiovascular Nursing Science Priorities: A Statement From the American Heart Association Council on Cardiovascular and Stroke Nursing.

The American Heart Association's (AHA) Council on Cardiovascular and Stroke Nursing (CVSN) plays a critical role in advancing the mission of the AHA in the discovery of new scientific knowledge. The aim was to identify priority research topics that would promote and improve cardiovascular (CV) health, provide direction for the education of future nurse scientists, and serve as a resource and catalyst for federal and organizational funding priorities. A Qualtrics survey, which included 3 questions about priorities for CVSN nurse researchers, was sent to the CVSN Leadership Committee and all CVSN Fellows of the AHA (n = 208). Responses to the questions were reviewed for word repetitions, patterns, and concepts and were then organized into thematic areas. The thematic areas were reviewed within small groups at the November (2016) in-person CVSN leadership meeting. Seventy-three surveys were completed. Five thematic areas were identified and included (1) developing and testing interventions, (2) assessment and monitoring, (3) precision CV nursing care, (4) translational and implementation science, and (5) big data. Topic areas noted were stroke, research methods, prevention of stroke and CV disease, self-management, and care and health disparities. Five thematic areas and 24 topic areas were identified as priorities for CV nursing research. These findings can provide a guide for CV nurse scientists and for federal and foundational funders to use in developing funding initiatives. We believe additional research and discovery in these thematic areas will help reduce the rising global burden of CV disease.

El estado del arte / estado de la técnica y la Investigación Científica y Tecnológica

El Estado del Arte/Estado de la técnica, hace referencia al estado último del conocimiento sobre la investigación y el desarrollo (I+D), es decir que es el límite de conocimiento generado sobre un tema o problema de investigación científica y/o tecnológica, estableciendo hasta donde ha avanzado el mismo, cual es la frontera en un tiempo y espacio determinado.En el inicio del proceso de construcción y/o descubrimiento de conocimientos científicos mediante la investigación, sin duda alguna, los profesores se preguntan ¿El estado del arte/técnica, es una de esas ocurrencias de la gestión de la investigación (Gestores), para complicar el trabajo de los investigadores en la formulación de sus protocolos de I+D? ¿Es el afán de hacer más burocrático el acceso al financiamiento, especialmente si este, tiene como fuente el presupuesto universitario? De igual manera, los investigadores con escasa experiencia en investigación tienen la percepción que el estado del arte/técnica es lo mismo que Marco Teórico y como tal lo tratan. En aras de contribuir a un mejor discernimiento e internalización y la adecuada utilización de estos elementos fundamentales de todo proyecto de investigación científica y/o tecnológica hemos preparado una síntesis sobre el tema.

Considerations on Geospatial Big Data

Geospatial data, as a significant portion of big data, has recently gained the full attention of researchers. However, few researchers focus on the evolution of geospatial data and its scientific research methodologies. When entering into the big data era, fully understanding the changing research paradigm associated with geospatial data will definitely benefit future research on big data. In this paper, we look deep into these issues by examining the components and features of geospatial big data, reviewing relevant scientific research methodologies, and examining the evolving pattern of geospatial data in the scope of the four ‘science paradigms’. This paper proposes that geospatial big data has significantly shifted the scientific research methodology from ‘hypothesis to data’ to ‘data to questions’ and it is important to explore the generality of growing geospatial data ‘from bottom to top’. Particularly, four research areas that mostly reflect data-driven geospatial research are proposed: spatial correlation, spatial analytics, spatial visualization, and scientific knowledge discovery. It is also pointed out that privacy and quality issues of geospatial data may require more attention in the future. Also, some challenges and thoughts are raised for future discussion.

Scientific Knowledge Engineering: a conceptual delineation and overview of the state of the art

AbstractAs a community work, scientific contributions are usually built incrementally, involving some transformation, expansion or refutation of existing conceptual and propositional networks. As the body of knowledge increases, scientists concentrate more effort on ensuring that new hypotheses and observations are needed and consistent with previous findings. In this paper, we will characterize Knowledge Engineering as an important groundwork for structuring scientific knowledge. We argue that knowledge-based computational infrastructures can support researchers in organizing and making explicit the main aspects needed to make inferences or extract conclusions from an existing body of knowledge. This view is also comparatively built, contrasting it with alternatives for manipulating scientific knowledge, namely data-intensive approaches and the computational discovery of scientific knowledge. The current state of the art is presented with 22 knowledge representations and computational infrastructure implementations, with their main relevant properties analyzed and compared. Based on this review and on the theoretical foundations of Knowledge Engineering, a high level step-by-step approach for specifying and constructing scientific computational environments is described. The paper concludes by indicating paths for further development of the view initiated here, especially related to the technical specificities that originates from applying Knowledge Engineering to scientific knowledge.

Featured Article: Genotation: Actionable knowledge for the scientific reader.

We present an article viewer application that allows a scientific reader to easily discover and share knowledge by linking genomics-related concepts to knowledge of disparate biomedical databases. High-throughput data streams generated by technical advancements have contributed to scientific knowledge discovery at an unprecedented rate. Biomedical Informaticists have created a diverse set of databases to store and retrieve the discovered knowledge. The diversity and abundance of such resources present biomedical researchers a challenge with knowledge discovery. These challenges highlight a need for a better informatics solution. We use a text mining algorithm, Genomine, to identify gene symbols from the text of a journal article. The identified symbols are supplemented with information from the GenoDB knowledgebase. Self-updating GenoDB contains information from NCBI Gene, Clinvar, Medgen, dbSNP, KEGG, PharmGKB, Uniprot, and Hugo Gene databases. The journal viewer is a web application accessible via a web browser. The features described herein are accessible on www.genotation.org The Genomine algorithm identifies gene symbols with an accuracy shown by .65 F-Score. GenoDB currently contains information regarding 59,905 gene symbols, 5633 drug-gene relationships, 5981 gene-disease relationships, and 713 pathways. This application provides scientific readers with actionable knowledge related to concepts of a manuscript. The reader will be able to save and share supplements to be visualized in a graphical manner. This provides convenient access to details of complex biological phenomena, enabling biomedical researchers to generate novel hypothesis to further our knowledge in human health. This manuscript presents a novel application that integrates genomic, proteomic, and pharmacogenomic information to supplement content of a biomedical manuscript and enable readers to automatically discover actionable knowledge.

Supporting scientific knowledge discovery with extended, generalized Formal Concept Analysis

In this paper we fuse together the Landscapes of Knowledge of Wille’s and Exploratory Data Analysis by leveraging Formal Concept Analysis (FCA) to support data-induced scientific enquiry and discovery.We use extended FCA first by allowing K-valued entries in the incidence to accommodate other, non-binary types of data, and second with different modes of creating formal concepts to accommodate diverse conceptualizing phenomena.With these extensions we demonstrate the versatility of the Landscapes of Knowledge metaphor to help in creating new scientific and engineering knowledge by providing several successful use cases of our techniques that support scientific hypothesis-making and discovery in a range of domains: semiring theory, perceptual studies, natural language semantics, and gene expression data analysis.While doing so, we also capture the affordances that justify the use of FCA and its extensions in scientific discovery.

K‐12 STEM Outreach: Ushering in the Next Generation of Scientists

Students likely to major in and earn college degrees in Science, Technology, Engineering, and Math (STEM) fields are those whose curiosity about STEM careers is piqued at an early age. At the K–12 level, this interest can be enhanced through hands‐on learning, projects with real life context and relevance, and contact with mentors working in these fields.The Interdisciplinary Science and Research (ISR) program at Stratford STEM Magnet High School in Nashville, TN has capitalized on a unique school/university partnership that utilizes a co‐teaching model pairing high school science teachers with Vanderbilt University scientists. The primary teaching strategy is inquiry‐based, incorporating multiple science and math disciplines into the coursework, promoting a facilitated learning model that relies heavily on the scientific method. In a school where the population is 74% minority and 81% of the students are considered economically disadvantaged, ISR provides a rigorous elective to challenge students in STEM disciplines, utilizing project‐based, real world learning.Students in ISR have consistently shown gains when compared to their peers including a 2.7 point increase on the ACT. Additionally, 69% of these students scored Proficient or Advanced on Algebra I and 59% on Biology I end of course exams compared to 33% and 13% of their peers, respectively. When surveyed, students report gains in critical thinking skills, improved public speaking ability, and increased adeptness in working collaboratively. This data emphasizes the need at the K‐12 level for students to be exposed to STEM professionals enhancing STEM content knowledge. While many scientists focus solely on the advancement of discovery and understanding of scientific knowledge, the value of sharing this knowledge with K‐12 students cannot be ignored.

Analysis on international scientific collaboration and research focus on depression field.

Depression is one of the most prevalent psychiatric diseases and significantly negatively impacts patients’ productivity, quality of life, and cognitive functions.[1] According to a recent survey by the World Health Organization, it has been estimated that depression currently ranks the third in terms of disability-adjusted life years and will be the leading cause of disease and injury burden by 2030, surpassing cancer as well as cardiovascular and respiratory disease.[2] This troubling situation, which has raised concern among the international field of psychiatry, challenges researchers worldwide to investigate the prevention and control of depression and identify the necessary funding to support this research. Given the expansive and interdisciplinary nature of biomedicine, no single researcher or institution can stay current on all developments within depression research. Thus, scientific collaboration becomes indispensable to make progress in biomedicine as it strengthens communication, encourages competency sharing, and facilitates the discovery of new scientific knowledge. Despite this fact, there have been few scientific publications to date regarding collaborative research within the field studying depression. Therefore, this article aimed to measure scientific collaborations at three levels – authors, institutions, and countries. By examining scientific collaboration, this article also identified hot topics in depression research, which represent the areas of focus that numerous researchers have decided to have substantial potential to prevent and cure this pervasive disease.

Soft infrastructure challenges to scientific knowledge discovery

Seeking to overcome nontechnical challenges to the scientific enterprise.