ABSTRACT
Background: COVID-19 is an infectious disease caused by SARS-CoV-2. The symptoms of COVID-19 vary from mild-to-moderate respiratory illnesses, and it sometimes requires urgent medication. Therefore, it is crucial to detect COVID-19 at an early stage through specific clinical tests, testing kits, and medical devices. However, these tests are not always available during the time of the pandemic. Therefore, this study developed an automatic, intelligent, rapid, and real-time diagnostic model for the early detection of COVID-19 based on its symptoms. Methods: The COVID-19 knowledge graph (KG) constructed based on literature from heterogeneous data is imported to understand the COVID-19 different relations. We added human disease ontology to the COVID-19 KG and applied a node-embedding graph algorithm called fast random projection to extract an extra feature from the COVID-19 dataset. Subsequently, experiments were conducted using two machine learning (ML) pipelines to predict COVID-19 infection from its symptoms. Additionally, automatic tuning of the model hyperparameters was adopted. Results: We compared two graph-based ML models, logistic regression (LR) and random forest (RF) models. The proposed graph-based RF model achieved a small error rate = 0.0064 and the best scores on all performance metrics, including specificity = 98.71%, accuracy = 99.36%, precision = 99.65%, recall = 99.53%, and F1-score = 99.59%. Furthermore, the Matthews correlation coefficient achieved by the RF model was higher than that of the LR model. Comparative analysis with other ML algorithms and with studies from the literature showed that the proposed RF model exhibited the best detection accuracy. Conclusion: The graph-based RF model registered high performance in classifying the symptoms of COVID-19 infection, thereby indicating that the graph data science, in conjunction with ML techniques, helps improve performance and accelerate innovations.
ABSTRACT
Disinformation spread on social media generates a truly massive amount of content on a daily basis, much of it not quite duplicated but repetitive and related. In this paper, we present an approach for clustering social media posts based on topic modeling in order to identify and formalize an underlying structure in all the noise. This would be of great benefit for tracking evolving trends, analyzing large-scale campaigns, and focusing efforts on debunking or community outreach. The steps we took in particular include harvesting through CrowdTangle huge collection of Facebook posts explicitly identified as containing disinformation by debunking experts, following those links back to the people, pages and groups where they were shared then collecting all posts shared on those channels over an extended period of time. This generated a very large textual dataset which was used in the topic modeling experiments attempting to identify the larger trends in the available data. Finally, the results were transformed and collected in a Knowledge Graph for further study and analysis. Our main goal is to investigate different trends and common patterns in disinformation campaigns, and whether there exist some correlations between some of them. For instance, for some of the most recent social media posts related to COVID-19 and political situation in Ukraine. © 2022 Copyright for this paper by its authors.
ABSTRACT
The outbreak of Covid-19 has exposed the lack of medical resources, especially the lack of medical personnel. This results in time and space restrictions for medical services, and patients cannot obtain health information all the time and everywhere. Based on the medical knowledge graph, healthcare bots alleviate this burden effectively by providing patients with diagnosis guidance, pre-diagnosis, and post-diagnosis consultation services in the way of human-machine dialogue. However, the medical utterance is more complicated in language structure, and there are complex intention phenomena in semantics. It is a challenge to detect the single intent, multi-intent, and implicit intent of a patient's utterance. To this end, we create a high-quality annotated Chinese Medical query (utterance) dataset, CMedQ (about 16.8k queries in medical domain which includes single, multiple, and implicit intents). It is hard to detect intent on such a complex dataset through traditional text classification models. Thus, we propose a novel detect model Conco-ERNIE, using concept co-occurrence patterns to enhance the representation of pre-trained model ERNIE. These patterns are mined using Apriori algorithm and will be embedded via Node2Vec. Their features will be aggregated with semantic features into Conco-ERNIE by using an attention module, which can catch user explicit intents and also predict user implicit intents. Experiments on CMedQ demonstrates that Conco-ERNIE achieves outstanding performance over baseline. Based on Conco-ERNIE, we develop an intelligent healthcare bot, MedicalBot. To provide knowledge support for MedicalBot, we also build a Chinese medical graph, CMedKG (about 45k entities and 283k relationships). © 2023 Association for Computing Machinery.
ABSTRACT
Public interest in where food comes from and how it is produced, processed, and distributed has increased over the last few decades, with even greater focus emerging during the COVID-19 pandemic. Mounting evidence and experience point to disturbing weaknesses in our food systems' abilities to support human livelihoods and wellbeing, and alarming long-term trends regarding both the environmental footprint of food systems and mounting vulnerabilities to shocks and stressors. How can we tackle the "wicked problems” embedded in a food system? More specifically, how can convergent research programs be designed and resulting knowledge implemented to increase inclusion, sustainability, and resilience within these complex systems, support widespread contributions to and acceptance of solutions to these challenges, and provide concrete benchmarks to measure progress and understand tradeoffs among strategies along multiple dimensions? This article introduces and defines food systems informatics (FSI) as a tool to enhance equity, sustainability, and resilience of food systems through collaborative, user-driven interaction, negotiation, experimentation, and innovation within food systems. Specific benefits we foresee in further development of FSI platforms include the creation of capacity-enabling verifiable claims of sustainability, food safety, and human health benefits relevant to particular locations and products;the creation of better incentives for the adoption of more sustainable land use practices and for the creation of more diverse agro-ecosystems;the wide-spread use of improved and verifiable metrics of sustainability, resilience, and health benefits;and improved human health through better diets.
ABSTRACT
Because of COVID-19 worldwide pandemic, there is a need for any complementary solutions to boost the immune system. Nowadays, healthy lifestyle, fitness, and diet habits have become central applications in our daily life. We designed a Naturopathy Knowledge Graph for a recommender system to boost the immune system (KISS: Knowledge-based Immune System Suggestion). The Naturopathy Knowledge Graph is built from more than 50 ontology-based food projects, also released as the LOV4IoT-Food ontology catalog. The naturopathy data set is referenced on the Linked Open Data (LOD) cloud. The LOV4IoT-Food ontology catalog encourages researchers to follow FAIR principles and share their reproducible experiments by publishing online their ontologies, data sets, rules, etc. The set of the ontology code shared online can be semiautomatically processed, if not available, the scientific publications describing the food ontologies are semiautomatically processed with Natural Language Processing (NLP) techniques. We build the naturopathy recommender system that will suggest food to boost the immune system. The recommender system can be extended to address other advice such as aromatherapy and take into consideration medical devices to monitor patients' vital signals. © 2022 Elsevier Inc. All rights reserved.
ABSTRACT
During the COVID-19 pandemic, researchers started to develop technical approaches to solve the numerous challenges imposed by the new pandemic. One fundamental precondition for research is to make relevant data about the COVID-19 pandemic available in a machine-processable way. For this purpose, COVID-19 ontologies and knowledge graphs have been developed and proposed for many different subareas of COVID-19 applications and research. In this paper, we provide a short analysis of the impact of COVID-19 ontologies. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
ABSTRACT
The healthcare industry faces many challenges like demand for high-quality remote services, especially when pandemics like COVID-19 spread across a region or even all over the world. Due to these challenges, healthcare providers are adopting innovative technologies to build new systems with enhanced automation for disease detection and assistance. For instance, a system able to support medical doctors to detect potential diseases when analyzing symptoms of a patient can help to treat the patient in a quicker and more effective manner, e.g., by routing her/him to the right specialist. Diagnostic systems need a significant amount of background knowledge in the medical sector, which can be enhanced by using semantics for knowledge representation, sharing, information integration and extraction, and reasoning. To this purpose, we propose a knowledge graph for medical diagnosis leveraging existing largely used standards and ontologies and we present the main issues in aligning them. Then we describe some usage scenarios for the knowledge graph. In detail, the knowledge graph for medical diagnosis encompasses SNOMED CT, ICD-10-CM, and DOID ontology. © 2022 Elsevier Inc. All rights reserved.
ABSTRACT
Medical dialog systems have the potential to assist e-medicine in improving access to healthcare services, improving patient treatment quality, and lowering medical expenses. In this research, we describe a knowledge-grounded conversation generation model that demonstrates how large-scale medical information in the form of knowledge graphs can aid in language comprehension and generation in medical dialog systems. Generic responses are often produced by existing generative dialog systems, resulting in monotonous and uninteresting conversations. To solve this problem, we combine various pre-trained language models with a medical knowledge base (UMLS) to generate clinically correct and human-like medical conversations using the recently released MedDialog-EN dataset. The medical-specific knowledge graph contains broadly 3 types of medical-related information, including disease, symptom and laboratory test. We perform reasoning over the retrieved knowledge graph by reading the triples in each graph using MedFact attention, which allows us to use semantic information from the graphs for better response generation. In order to preserve medical information, we employ a policy network, which effectively injects relevant entities associated with each dialog into the response. We also study how transfer learning can significantly improve the performance by utilizing a relatively small corpus, created by extending the recently released CovidDialog dataset, containing the dialogs for diseases that are symptoms of Covid-19. Empirical results on the MedDialog corpus and the extended CovidDialog dataset demonstrate that our proposed model significantly outperforms the state-of-the-art methods in terms of both automatic evaluation and human judgment.
Subject(s)
COVID-19 , Pattern Recognition, Automated , Humans , Semantics , Unified Medical Language System , CommunicationABSTRACT
Question Answering based on Knowledge Graph (KG) has emerged as a popular research area in general domain. However, few works focus on the COVID-19 kg-based question answering, which is very valuable for biomedical domain. In addition, existing question answering methods rely on knowledge embedding models to represent knowledge (i.e., entities and questions), but the relations between entities are neglected. In this paper, we construct a COVID-19 knowledge graph and propose an end-to-end knowledge graph question answering approach that can utilize relation information to improve the performance. Experimental result shows that the effectiveness of our approach on the COVID-19 knowledge graph question answering. Our code and data are available at https://github.com/CHNcreater/COVID-19-KGQA. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
ABSTRACT
A knowledge graph (KG) consists of numerous triples, in which each triple, i.e., (head entity, relation, tail entity), denotes a real-world assertion. Many large-scale KGs have been developed, e.g., general-purpose KGs Freebase and YAGO. Also, lots of domain-specific KGs are emerging, e.g., COVID-19 KGs, biomedical KGs, and agricultural KGs. By embedding KGs into low-dimensional vectors, i.e., representations of entities and relations, we could integrate KGs into machine learning models and enhance the performance of many prediction tasks, including search, recommendations, and question answering. During the construction, refinement, embedding, and application of KGs, a variety of KG learning algorithms have been developed to handle challenges in various real-world scenarios. Moreover, graph neural networks have also brought new opportunities to KG learning. This workshop aims to engage with active researchers from KG communities, recommendation communities, natural language processing communities, and other communities, and deliver state-of-the-art research insights into the core challenges in KG learning. © 2023 Owner/Author.
ABSTRACT
This paper presents a computational approach designed to construct and query a literature-based knowledge graph for predicting novel drug therapeutics. The main objective is to offer a platform that discovers drug combinations from FDA-approved drugs and accelerates their investigations by domain scientists. Specifically, the paper introduced the following algorithms: (1) an algorithm for constructing the knowledge graph from drug, gene, and disease mentions in the biomedical literature;(2) an algorithm for vetting the knowledge graph from drug combinations that may pose a risk of drug interaction;(3) and two querying algorithms for searching the knowledge graph by a single drug or a combination of drugs. The resulting knowledge graph had 844 drugs, 306 gene/protein features, and 19 disease mentions. The original number of drug combinations generated was 2,001. We queried the knowledge graph to eliminate noise generated from chemicals that are not drugs. This step resulted in 614 drug combinations. When vetting the knowledge graph to eliminate the potentially risky drug combinations, it resulted in predicting 200 combinations. Our domain expert manually eliminated extra 54 combinations which left only 146 combination candidates. Our three-layered knowledge graph, empowered by our algorithms, offered a tool that predicted drug combination therapeutics for scientists who can further investigate from the viewpoint of drug targets and side effects. © 2022 IEEE.
ABSTRACT
Fighting against misinformation and computational propaganda requires integrated efforts from various domains like law or education, but there is also a need for computational tools. I investigate here how reasoning in Description Logics (DLs) can detect inconsistencies between trusted knowledge and not trusted sources. The proposed method is exemplified on fake news for the new coronavirus. Indeed, in the context of the Covid-19 pandemic, many were quick to spread deceptive information. Since, the not-trusted information comes in natural language (e.g. "Covid-19 affects only the elderly”), the natural language text is automatically converted into DLs using the FRED tool. The resulted knowledge graph formalised in Description Logics is merged with the trusted ontologies on Covid-10. Reasoning in Description Logics is then performed with the Racer reasoner, which is responsable to detect inconsistencies within the ontology. When detecting inconsistencies, a "red flag” is raised to signal possible fake news. The reasoner can provide justifications for the detected inconsistency. This availability of justifications is the main advantage compared to approaches based on machine learning, since the system is able to explain its reasoning steps to a human agent. Hence, the approach is a step towards human-centric AI systems. The main challenge remains to improve the technology which automatically translates text into some formal representation. © 2022, Springer Nature Switzerland AG.
ABSTRACT
To combat COVID-19, scientists must digest the vast amount of relevant biomedical knowledge in the literature to understand disease mechanisms and related biological functions. Nearly 3,000 scientific papers are published on PubMed every day. This knowledge bottleneck has resulted in severe delays in developing COVID-19 vaccines and drugs. Our research produces a hierarchy of knowledge concepts related to COVID-19, designed to assist scientists in answering questions and generating summaries. It aims to discover scientific and comprehensive knowledge to extract fine-grained multimedia elements (i.e., physical and visual structures, relational events and events, and chemical knowledge). Our project is toward one step in natural language understanding: detailed contextual sentences, subgraphs, and knowledge subgraphs are the first time to be automatically generated, and relations and coreferences of COVID-19 mentions will be sketched. Extensive results show that our method outperforms other state-of-the-art methods. In addition, we have published the generated knowledge graph on Google Drive1 and released the source in the Github2. © 2022 ACM.
ABSTRACT
The medical conversational system can relieve doctors' burden and improve healthcare effi-ciency, especially during the COVID-19 pan-demic. However, the existing medical dialogue systems have die problems of weak scalability, insufficient knowledge, and poor controlla-bility. Thus, we propose a medical conversa-tional question-answering (CQA) system based on the knowledge graph, namely MedConQA, which is designed as a pipeline framework to maintain high flexibility. Our system utilizes automated medical procedures, including medi-cal triage, consultation, image-text drug recom-mendation, and record. Each module has been open-sourced as a tool, which can be used alone or in combination, with robust scalability. Besides, to conduct knowledge-grounded dia-logues with users, we first construct a Chinese Medical Knowledge Graph (CMKG) and col-lect a large-scale Chinese Medical CQA (CM-CQA) dataset, and we design a series of meth-ods for reasoning more intellectually. Finally, we use several state-of-the-art (SOTA) tech-niques to keep the final generated response more controllable, which is further assured by hospital and professional evaluations. We have open-sourced related code, datasets, web pages, and tools, hoping to advance future research. © 2022 Association for Computational Linguistics.
ABSTRACT
The proceedings contain 27 papers. The topics discussed include: UKP-SQUARE: an online platform for question answering research;ViLMedic: a framework for research at the intersection of vision and language in medical AI;TextPruner: a model pruning toolkit for pre-trained language models;AnnIE: an annotation platform for constructing complete open information extraction benchmark;AdapterHub playground: simple and flexible few-shot learning with adapters;QiuNiu: a Chinese lyrics generation system with passage-level input;automatic gloss dictionary for sign language learners;PromptSource: an integrated development environment and repository for natural language prompts;COVID-19 claim radar: a structured claim extraction and tracking system;TS-Anno: an annotation tool to build, annotate and evaluate text simplification corpora;and CogKGE: a knowledge graph embedding toolkit and benchmark for representing multi-source and heterogeneous knowledge.
ABSTRACT
Identifying effective drug treatments for COVID-19 is essential to reduce morbidity and mortality. Although a number of existing drugs have been proposed as potential COVID-19 treatments, effective data platforms and algorithms to prioritize drug candidates for evaluation and application of knowledge graph for drug repurposing have not been adequately explored. A COVID-19 knowledge graph by integrating 14 public bioinformatic databases containing information on drugs, genes, proteins, viruses, diseases, symptoms and their linkages is developed. An algorithm is developed to extract hidden linkages connecting drugs and COVID-19 from the knowledge graph, to generate and rank proposed drug candidates for repurposing as treatments for COVID-19 by integrating three scores for each drug: motif scores, knowledge graph PageRank scores, and knowledge graph embedding scores. The knowledge graph contains over 48 000 nodes and 13 37 000 edges, including 13 563 molecules in the DrugBank database. From the 5624 molecules identified by the motif-discovery algorithms, ranking results show that 112 drug molecules had the top 2% scores, of which 50 existing drugs with other indications approved by health administrations reported. The proposed drug candidates serve to generate hypotheses for future evaluation in clinical trials and observational studies.
ABSTRACT
Recent advances in the healthcare industry have led to an abundance of unstructured data, making it challenging to perform tasks such as efficient and accurate information retrieval at scale. Our work offers an all-in-one scalable solution for extracting and exploring complex information from large-scale research documents, which would otherwise be tedious. First, we briefly explain our knowledge synthesis process to extract helpful information from unstructured text data of research documents. Then, on top of the knowledge extracted from the documents, we perform complex information retrieval using three major components- Paragraph Retrieval, Triplet Retrieval from Knowledge Graphs, and Complex Question Answering (QA). These components combine lexical and semantic-based methods to retrieve paragraphs and triplets and perform faceted refinement for filtering these search results. The complexity of biomedical queries and documents necessitates using a QA system capable of handling queries more complex than factoid queries, which we evaluate qualitatively on the COVID-19 Open Research Dataset (CORD-19) to demonstrate the effectiveness and value-add. © 2022 IEEE.
ABSTRACT
Since COVID-19 emerged in 2019, significant levels of suffering and disruption have been caused on a global scale. Although vaccines have become widely used, the virus has shown its potential for evading immunities or acquiring other novel characteristics. Whether current drug treatments are still effective for people infected with Omicron remains unclear. Due to the long development cycles and high expense requirements of de novo drug development, many researchers have turned to consider drug repositioning in the search to find effective treatments for COVID-19. Here, we review such drug repositioning and combination efforts towards providing better handling. For potential drugs under consideration, aspects of both structure and function require attention, with specific categories of sequence, expression, structure, and interaction, the key parameters for investigation. For different data types, we show the corresponding differing drug repositioning methods that have been exploited. As incorporating drug combinations can increase therapeutic efficacy and reduce toxicity, we also review computational strategies to reveal drug combination potential. Taken together, we found that graph theory and neural network were the most used strategy with high potential towards drug repositioning for COVID-19. Integrating different levels of data may further improve the success rate of drug repositioning.
ABSTRACT
Introduction: Tracing transmission paths and identifying infection sources have been effective in curbing the spread of coronavirus disease 2019 (COVID-19). However, when facing a large-scale outbreak, this is extremely time-consuming and labor-intensive, and resources for infection source tracing become limited. In this study, we aimed to use knowledge graph (KG) technology to automatically infer transmission paths and infection sources. Methods: We constructed a KG model to automatically extract epidemiological information and contact relationships from case reports. We then used an inference engine to identify transmission paths and infection sources. To test the model's performance, we used data from two COVID-19 outbreaks in Beijing. Results: The KG model performed well for both outbreaks. In the first outbreak, 20 infection relationships were identified manually, while 42 relationships were determined using the KG model. In the second outbreak, 32 relationships were identified manually and 31 relationships were determined using the KG model. All discrepancies and omissions were reasonable. Discussion: The KG model is a promising tool for predicting and controlling future COVID-19 epidemic waves and other infectious disease pandemics. By automatically inferring the source of infection, limited resources can be used efficiently to detect potential risks, allowing for rapid outbreak control.
ABSTRACT
AI-driven approaches are widely used in drug discovery, where candidate molecules are generated and tested on a target protein for binding affinity prediction. However, generating new compounds with desirable molecular properties such as Quantitative Estimate of Drug-likeness (QED) and Dopamine Receptor D2 activity (DRD2) while adhering to distinct chemical laws is challenging. To address these challenges, we proposed a graph-based deep learning framework to generate potential therapeutic drugs targeting the SARS-CoV-2 protein. Our proposed framework consists of two modules: a novel reinforcement learning (RL)-based graph generative module with knowledge graph (KG) and a graph early fusion approach (GEFA) for binding affinity prediction. The first module uses a gated graph neural network (GGNN) model under the RL environment for generating novel molecular compounds with desired properties and a custom-made KG for molecule screening. The second module uses GEFA to predict binding affinity scores between the generated compounds and target proteins. Experiments show how fine-tuning the GGNN model under the RL environment enhances the molecules with desired properties to generate 100 % valid and 100 % unique compounds using different scoring functions. Additionally, KG-based screening reduces the search space of generated candidate molecules by 96.64 % while retaining 95.38 % of promising binding molecules against SARS-CoV-2 protein, i.e., 3C-like protease (3CLpro). We achieved a binding affinity score of 8.185 from the top rank of generated compound. In addition, we compared top-ranked generated compounds to Indinavir on different parameters, including drug-likeness and medicinal chemistry, for qualitative analysis from a drug development perspective. Supplementary Information: The online version contains supplementary material available at 10.1007/s13721-023-00409-2.