A BERT-based ensemble learning approach for the BioCreative VII challenges: full-text chemical identification and multi-label classification in PubMed articles.

In this research, we explored various state-of-the-art biomedical-specific pre-trained Bidirectional Encoder Representations from Transformers (BERT) models for the National Library of Medicine - Chemistry (NLM CHEM) and LitCovid tracks in the BioCreative VII Challenge, and propose a BERT-based ensemble learning approach to integrate the advantages of various models to improve the system's performance. The experimental results of the NLM-CHEM track demonstrate that our method can achieve remarkable performance, with F1-scores of 85% and 91.8% in strict and approximate evaluations, respectively. Moreover, the proposed Medical Subject Headings identifier (MeSH ID) normalization algorithm is effective in entity normalization, which achieved a F1-score of about 80% in both strict and approximate evaluations. For the LitCovid track, the proposed method is also effective in detecting topics in the Coronavirus disease 2019 (COVID-19) literature, which outperformed the compared methods and achieve state-of-the-art performance in the LitCovid corpus. Database URL: https://www.ncbi.nlm.nih.gov/research/coronavirus/.

Multi-probe attention neural network for COVID-19 semantic indexing.

BACKGROUND: The COVID-19 pandemic has increasingly accelerated the publication pace of scientific literature. How to efficiently curate and index this large amount of biomedical literature under the current crisis is of great importance. Previous literature indexing is mainly performed by human experts using Medical Subject Headings (MeSH), which is labor-intensive and time-consuming. Therefore, to alleviate the expensive time consumption and monetary cost, there is an urgent need for automatic semantic indexing technologies for the emerging COVID-19 domain. RESULTS: In this research, to investigate the semantic indexing problem for COVID-19, we first construct the new COVID-19 Semantic Indexing dataset, which consists of more than 80 thousand biomedical articles. We then propose a novel semantic indexing framework based on the multi-probe attention neural network (MPANN) to address the COVID-19 semantic indexing problem. Specifically, we employ a k-nearest neighbour based MeSH masking approach to generate candidate topic terms for each input article. We encode and feed the selected candidate terms as well as other contextual information as probes into the downstream attention-based neural network. Each semantic probe carries specific aspects of biomedical knowledge and provides informatively discriminative features for the input article. After extracting the semantic features at both term-level and document-level through the attention-based neural network, MPANN adopts a linear multi-view classifier to conduct the final topic prediction for COVID-19 semantic indexing. CONCLUSION: The experimental results suggest that MPANN promises to represent the semantic features of biomedical texts and is effective in predicting semantic topics for COVID-19 related biomedical articles.

A new perspective: Practice-enhancing publications about the medication-use process in ambulatory care in 2020.

PURPOSE: This article identifies, prioritizes, and summarizes published literature on the ambulatory care medication-use process (ACMUP) from calendar year 2020 that can impact ambulatory pharmacy practice. SUMMARY: The medication-use process is the foundational system that provides the framework for safe medication utilization within the healthcare environment and was reimagined to focus on new innovations and advancements in ambulatory pharmacy practice. The ACMUP is defined in this article as having the following components: transitions of care, prescribing and collaborative practice, accessing care, adherence, and monitoring and quality. Articles evaluating at least one step of the ACMUP were assessed for their usefulness toward practice improvement. A PubMed search covering calendar year 2020 was conducted in January 2021 using targeted Medical Subject Headings (MeSH) keywords and the table of contents of selected pharmacy journals, providing a total of 9,433 articles. A thorough review identified 65 potentially practice-enhancing articles: 14 for transitions of care, 19 for prescribing and collaborative practice, 10 for adherence, 6 for accessing care, and 16 for monitoring and quality. Ranking of the articles for importance by peers led to the selection of key articles from each category. The highest-ranked articles are briefly summarized, with a mention of why each article is important. The other articles are listed for further review and evaluation. CONCLUSION: It is important to routinely review the published literature and to incorporate significant findings into daily practice. This article is the first to define and evaluate the currently published literature pertinent to the ACMUP. As healthcare continues to advance and care shifts to ambulatory settings, the ACMUP will continue to be a crucial process to evaluate.

The impact of the COVID-19 pandemic on scientific research in the life sciences.

The COVID-19 outbreak has posed an unprecedented challenge to humanity and science. On the one side, public and private incentives have been put in place to promptly allocate resources toward research areas strictly related to the COVID-19 emergency. However, research in many fields not directly related to the pandemic has been displaced. In this paper, we assess the impact of COVID-19 on world scientific production in the life sciences and find indications that the usage of medical subject headings (MeSH) has changed following the outbreak. We estimate through a difference-in-differences approach the impact of the start of the COVID-19 pandemic on scientific production using the PubMed database (3.6 Million research papers). We find that COVID-19-related MeSH terms have experienced a 6.5 fold increase in output on average, while publications on unrelated MeSH terms dropped by 10 to 12%. The publication weighted impact has an even more pronounced negative effect (-16% to -19%). Moreover, COVID-19 has displaced clinical trial publications (-24%) and diverted grants from research areas not closely related to COVID-19. Note that since COVID-19 publications may have been fast-tracked, the sudden surge in COVID-19 publications might be driven by editorial policy.

A Bibliometric Analysis of 14,822 Researches on Myocardial Reperfusion Injury by Machine Learning.

Myocardial ischemia is the major cause of death worldwide, and reperfusion is the standard intervention for myocardial ischemia. However, reperfusion may cause additional damage, known as myocardial reperfusion injury, for which there is still no effective therapy. This study aims to analyze the landscape of researches concerning myocardial reperfusion injury over the past three decades by machine learning. PubMed was searched for publications from 1990 to 2020 indexed under the Medical Subject Headings (MeSH) term "myocardial reperfusion injury" on 13 April 2021. MeSH analysis and Latent Dirichlet allocation (LDA) analyses were applied to reveal research hotspots. In total, 14,822 publications were collected and analyzed in this study. MeSH analyses revealed that time factors and apoptosis were the leading terms of the pathogenesis and treatment of myocardial reperfusion injury, respectively. In LDA analyses, research topics were classified into three clusters. Complex correlations were observed between topics of different clusters, and the prognosis is the most concerned field of the researchers. In conclusion, the number of publications on myocardial reperfusion injury increases during the past three decades, which mainly focused on prognosis, mechanism, and treatment. Prognosis is the most concerned field, whereas studies on mechanism and treatment are relatively lacking.

NewsMeSH: A new classifier designed to annotate health news with MeSH headings.

MOTIVATION: In the age of big data, the amount of scientific information available online dwarfs the ability of current tools to support researchers in locating and securing access to the necessary materials. Well-structured open data and the smart systems that make the appropriate use of it are invaluable and can help health researchers and professionals to find the appropriate information by, e.g., configuring the monitoring of information or refining a specific query on a disease. METHODS: We present an automated text classifier approach based on the MEDLINE/MeSH thesaurus, trained on the manual annotation of more than 26 million expert-annotated scientific abstracts. The classifier was developed tailor-fit to the public health and health research domain experts, in the light of their specific challenges and needs. We have applied the proposed methodology on three specific health domains: the Coronavirus, Mental Health and Diabetes, considering the pertinence of the first, and the known relations with the other two health topics. RESULTS: A classifier is trained on the MEDLINE dataset that can automatically annotate text, such as scientific articles, news articles or medical reports with relevant concepts from the MeSH thesaurus. CONCLUSIONS: The proposed text classifier shows promising results in the evaluation of health-related news. The application of the developed classifier enables the exploration of news and extraction of health-related insights, based on the MeSH thesaurus, through a similar workflow as in the usage of PubMed, with which most health researchers are familiar.

Searching PubMed to Retrieve Publications on the COVID-19 Pandemic: Comparative Analysis of Search Strings.

BACKGROUND: Since it was declared a pandemic on March 11, 2020, COVID-19 has dominated headlines around the world and researchers have generated thousands of scientific articles about the disease. The fast speed of publication has challenged researchers and other stakeholders to keep up with the volume of published articles. To search the literature effectively, researchers use databases such as PubMed. OBJECTIVE: The aim of this study is to evaluate the performance of different searches for COVID-19 records in PubMed and to assess the complexity of searches required. METHODS: We tested PubMed searches for COVID-19 to identify which search string performed best according to standard metrics (sensitivity, precision, and F-score). We evaluated the performance of 8 different searches in PubMed during the first 10 weeks of the COVID-19 pandemic to investigate how complex a search string is needed. We also tested omitting hyphens and space characters as well as applying quotation marks. RESULTS: The two most comprehensive search strings combining several free-text and indexed search terms performed best in terms of sensitivity (98.4%/98.7%) and F-score (96.5%/95.7%), but the single-term search COVID-19 performed best in terms of precision (95.3%) and well in terms of sensitivity (94.4%) and F-score (94.8%). The term Wuhan virus performed the worst: 7.7% for sensitivity, 78.1% for precision, and 14.0% for F-score. We found that deleting a hyphen or space character could omit a substantial number of records, especially when searching with SARS-CoV-2 as a single term. CONCLUSIONS: Comprehensive search strings combining free-text and indexed search terms performed better than single-term searches in PubMed, but not by a large margin compared to the single term COVID-19. For everyday searches, certain single-term searches that are entered correctly are probably sufficient, whereas more comprehensive searches should be used for systematic reviews. Still, we suggest additional measures that the US National Library of Medicine could take to support all PubMed users in searching the COVID-19 literature.

Analysis of COVID-19 clinical trials: A data-driven, ontology-based, and natural language processing approach.

With the novel COVID-19 pandemic disrupting and threatening the lives of millions, researchers and clinicians have been recently conducting clinical trials at an unprecedented rate to learn more about the virus and potential drugs/treatments/vaccines to treat its infection. As a result of the influx of clinical trials, researchers, clinicians, and the lay public, now more than ever, face a significant challenge in keeping up-to-date with the rapid rate of discoveries and advances. To remedy this problem, this research mined the ClinicalTrials.gov corpus to extract COVID-19 related clinical trials, produce unique reports to summarize findings and make the meta-data available via Application Programming Interfaces (APIs). Unique reports were created for each drug/intervention, Medical Subject Heading (MeSH) term, and Human Phenotype Ontology (HPO) term. These reports, which have been run over multiple time points, along with APIs to access meta-data, are freely available at http://covidresearchtrials.com. The pipeline, reports, association of COVID-19 clinical trials with MeSH and HPO terms, insights, public repository, APIs, and correlations produced are all novel in this work. The freely available, novel resources present up-to-date relevant biological information and insights in a robust, accessible manner, illustrating their invaluable potential to aid researchers overcome COVID-19 and save hundreds of thousands of lives.