bims-librar Biomed News
on Biomedical librarianship
Issue of 2020‒08‒23
thirteen papers selected by
Thomas Krichel
Open Library Society

  1. JBI Evid Synth. 2020 Jul 06.
    Lieggi M, Olson L, Kleiman A, Jang H.
      OBJECTIVE: The objective of the review is to evaluate how health care providers working in hospitals perceive clinical librarian services.INTRODUCTION: Clinical librarian programs have existed IMO as early as 1971, however, there is a lack of evidence on their effectiveness in affecting health care outcomes. Studies report primarily on programs supporting medicine, although these programs also support other health care providers. In order for clinical librarians to affect outcomes, particularly those focused on patient-centered, evidence-based care, they need insight into how hospital health care providers perceive clinical librarian services.
    INCLUSION CRITERIA: The review will consider studies that include any health care provider that works within a hospital, including surgical, clinical, and inpatient units. Studies that focus on qualitative data about clinical librarian services, published from 1971, will be eligible for inclusion.
    METHODS: The primary databases to be searched are PubMed, CINAHL, Embase, PsycINFO, Library Literature & Information Science, and LISTA (Library, Information Science & Technology Abstracts), and Web of Science. Studies will be selected based on their assessment against the inclusion criteria by two independent reviewers. Eligible studies will be critically appraised for methodological quality. Data will be extracted using a standardized tool and findings pooled and synthesized using a meta-aggregation approach.
    SYSTEMATIC REVIEW REGISTRATION NUMBER: We are rejecting this submission as we do not consider it to meet the scope of PROSPERO as reviews need to contain at least one outcome of direct patient or clinical relevance in order to be included in PROSPERO.
  2. J Biomed Inform. 2020 Aug 17. pii: S1532-0464(20)30158-1. [Epub ahead of print] 103530
    Rybinski M, Xu J, Karimi S.
      Bidirectional Encoder Representations from Transformers (BERT) have achieved state-of-the-art effectiveness in some of the biomedical information processing applications. We investigate the effectiveness of these techniques for clinical trial search systems. In precision medicine, matching patients to relevant experimental evidence or prospective treatments is a complex task which requires both clinical and biological knowledge. To assist in this complex decision making, we investigate the effectiveness of different ranking models based on the BERT models under the same retrieval platform to ensure fair comparisons. An evaluation on the TREC Precision Medicine benchmarks indicates that our approach using the BERT model pre-trained on scientific abstracts and clinical notes achieves state-of-the-art results, on par with highly specialised, manually optimised heuristic models. We also report the best results to date on the TREC Precision Medicine 2017 ad hoc retrieval task for clinical trial search.
    Keywords:  Bidirectional transformer encoder; Clinical decision making; Complex search; Document Search; Information retrieval; Learning-to-Rank; Natural language processing; Precision medicine; Ranking functions
  3. J Stroke Cerebrovasc Dis. 2020 Sep;pii: S1052-3057(20)30460-2. [Epub ahead of print]29(9): 105042
    Fischer I, Steiger HJ.
      BACKGROUND: Text mining with automatic extraction of key features is gaining increasing importance in science and particularly medicine due to the rapidly increasing number of publications.OBJECTIVES: Here we evaluate the current potential of sentiment analysis and machine learning to extract the importance of the reported results and conclusions of randomized trials on stroke.
    METHODS: PubMed abstracts of 200 recent reports of randomized trials were reviewed and manually classified according to the estimated importance of the studies. Importance of the papers was classified as "game changer", "suggestive", "maybe" "negative result". Algorithmic sentiment analysis was subsequently used on both the "Results" and the "Conclusions" paragraphs, resulting in a numerical output for polarity and subjectivity. The result of the human assessment was then compared to polarity and subjectivity. In addition, a neural network using the Keras platform built on Tensorflow and Python was trained to map the "Results" and "Conclusions" to the dichotomized human assessment (1: "game changer" or "suggestive"; 0:"maybe" or "negative", or no results reported). 120 abstracts were used as the training set and 80 as the test set.
    RESULTS: 9 out of the 200 reports were classified manually as "game changer", 40 as "suggestive", 73 as "maybe" and 32 and "negative"; 46 abstracts did not contain any results. Polarity was generally higher for the "Conclusions" than for the "Results". Polarity was highest for the "Conclusions" classified as "suggestive". Subjectivity was also higher in the classes "suggestive" and "maybe" than in the classes "game changer" and "negative". The trained neural network provided a correct dichotomized output with an accuracy of 71% based on the "Results" and 73% based on "Conclusions" .
    CONCLUSIONS: Current statistical approaches to text analysis can grasp the impact of scientific medical abstracts to a certain degree. Sentiment analysis showed that mediocre results are apparently written in more enthusiastic words than clearly positive or negative results.
    Keywords:  Artificial neural network; Machine learning; PubMed abstracts; Sentiment analysis; Text mining
  4. Patient Educ Couns. 2020 Aug 07. pii: S0738-3991(20)30402-X. [Epub ahead of print]
    Jayasinghe R, Ranasinghe S, Jayarajah U, Seneviratne S.
      OBJECTIVES: To analyse the quality of information included in websites aimed at the public on COVID-19.METHODS: Yahoo!, Google and Bing search engines were browsed using selected keywords on COVID-19. The first 100 websites from each search engine for each keyword were evaluated. Validated tools were used to assess readability [Flesch Reading Ease Score (FRES)], usability and reliability (LIDA tool) and quality (DISCERN instrument). Non-parametric tests were used for statistical analyses.
    RESULTS: Eighty-four eligible sites were analysed. The median FRES score was 54.2 (range: 23.2-73.5). The median LIDA usability and reliability scores were 46 (range: 18-54) and 37(range:14-51), respectively. A low (<50 %) overall LIDA score was recorded for 30.9 % (n = 26) of the websites. The median DISCERN score was 49.5 (range: 21-77). The DISCERN score of ≤50 % was found in 45 (53.6 %) websites. The DISCERN score was significantly associated with LIDA usability and reliability scores (p < 0.001) and the FRES score (p = 0.024).
    CONCLUSION: The majority of websites on COVID-19 for the public had moderate to low scores with regards to readability, usability, reliability and quality.
    PRACTICE IMPLICATIONS: Prompt strategies should be implemented to standardize online health information on COVID-19 during this pandemic to ensure the general public has access to good quality reliable information.
    Keywords:  2019-novel coronavirus; COVID-19; Education websites; General public; Internet; Quality; Quality of information; SARS-CoV-2
  5. ScientificWorldJournal. 2020 ;2020 1562028
    Joshi A, Kajal F, Bhuyan SS, Sharma P, Bhatt A, Kumar K, Kaur M, Arora A.
      Background: The novel coronavirus disease (COVID-19) has spread globally from its epicenter in Hubei, China, and was declared a pandemic by the World Health Organization (WHO) on March 11, 2020. The most popular search engine worldwide is Google, and since March 2020, COVID-19 has been a global trending search term. Misinformation related to COVID-19 from these searches is a problem, and hence, it is of high importance to assess the quality of health information over the internet related to COVID-19. The objective of our study is to examine the quality of COVID-19 related health information over the internet using the DISCERN tool.Methods: The keywords included in assessment of COVID-19 related information using Google's search engine were "Coronavirus," "Coronavirus causes," "Coronavirus diagnosis," "Coronavirus prevention," and "Coronavirus management". The first 20 websites from each search term were gathered to generate a list of 100 URLs. Duplicate sites were excluded from this search, allowing analysis of unique sites only. Additional exclusion criteria included scientific journals, nonoperational links, nonfunctional websites (where the page was not loading, was not found, or was inactive), and websites in languages other than English. This resulted in a unique list of 48 websites. Four independent raters evaluated the websites using a 16-item DISCERN tool to assess the quality of novel coronavirus related information available on the internet. The interrater reliability agreement was calculated using the intracluster correlation coefficient.
    Results: Results showed variation in how the raters assigned scores to different website categories. The .com websites received the lowest scores. Results showed that .edu and .org website category sites were excellent in communicating coronavirus related health information; however, they received lower scores for treatment effect and treatment choices.
    Conclusion: This study highlights the gaps in the quality of information that is available on the websites related to COVID-19 and study emphasizes the need for verified websites that provide evidence-based health information related to the novel coronavirus pandemic.
  6. Yearb Med Inform. 2020 Aug;29(1): 208-220
    Hahn U, Oleynik M.
      OBJECTIVES: We survey recent developments in medical Information Extraction (IE) as reported in the literature from the past three years. Our focus is on the fundamental methodological paradigm shift from standard Machine Learning (ML) techniques to Deep Neural Networks (DNNs). We describe applications of this new paradigm concentrating on two basic IE tasks, named entity recognition and relation extraction, for two selected semantic classes-diseases and drugs (or medications)-and relations between them.METHODS: For the time period from 2017 to early 2020, we searched for relevant publications from three major scientific communities: medicine and medical informatics, natural language processing, as well as neural networks and artificial intelligence.
    RESULTS: In the past decade, the field of Natural Language Processing (NLP) has undergone a profound methodological shift from symbolic to distributed representations based on the paradigm of Deep Learning (DL). Meanwhile, this trend is, although with some delay, also reflected in the medical NLP community. In the reporting period, overwhelming experimental evidence has been gathered, as illustrated in this survey for medical IE, that DL-based approaches outperform non-DL ones by often large margins. Still, small-sized and access-limited corpora create intrinsic problems for data-greedy DL as do special linguistic phenomena of medical sublanguages that have to be overcome by adaptive learning strategies.
    CONCLUSIONS: The paradigm shift from (feature-engineered) ML to DNNs changes the fundamental methodological rules of the game for medical NLP. This change is by no means restricted to medical IE but should also deeply influence other areas of medical informatics, either NLP- or non-NLP-based.
  7. Methods Mol Biol. 2021 ;2190 289-305
    Sousa D, Lamurias A, Couto FM.
      Using different sources of information to support automated extracting of relations between biomedical concepts contributes to the development of our understanding of biological systems. The primary comprehensive source of these relations is biomedical literature. Several relation extraction approaches have been proposed to identify relations between concepts in biomedical literature, namely, using neural networks algorithms. The use of multichannel architectures composed of multiple data representations, as in deep neural networks, is leading to state-of-the-art results. The right combination of data representations can eventually lead us to even higher evaluation scores in relation extraction tasks. Thus, biomedical ontologies play a fundamental role by providing semantic and ancestry information about an entity. The incorporation of biomedical ontologies has already been proved to enhance previous state-of-the-art results.
    Keywords:  Biomedical literature; Deep learning; External sources of knowledge; Neural networks; Ontologies; Relation extraction
  8. Yearb Med Inform. 2020 Aug;29(1): 221-225
    Grouin C, Grabar N, .
      OBJECTIVES: Analyze papers published in 2019 within the medical natural language processing (NLP) domain in order to select the best works of the field.METHODS: We performed an automatic and manual pre-selection of papers to be reviewed and finally selected the best NLP papers of the year. We also propose an analysis of the content of NLP publications in 2019.
    RESULTS: Three best papers have been selected this year including the generation of synthetic record texts in Chinese, a method to identify contradictions in the literature, and the BioBERT word representation.
    CONCLUSIONS: The year 2019 was very rich and various NLP issues and topics were addressed by research teams. This shows the will and capacity of researchers to move towards robust and reproducible results. Researchers also prove to be creative in addressing original issues with relevant approaches.
  9. J Dent Educ. 2020 Aug 19.
    Cesur Aydin K, Gunec HG.
      OBJECTIVES: This study was designed to investigate Artificial Intelligence in Dental Radiology (AIDR) videos on YouTube in terms of popularity, content, reliability, and educational quality.METHODS: Two researchers systematically searched about AIDR on YouTube on January 27, 2020, by using the terms "artificial intelligence in dental radiology," "machine learning in dental radiology," and "deep learning in dental radiology." The search was performed in English, and 60 videos for each keyword were assessed. Video source, content type, time since upload, duration, and number of views, likes, and dislikes were recorded. Video popularity was reported using Video Power Index (VPI). The accuracy and reliability of the source of information were measured using the adapted DISCERN score. The quality of the videos was measured using JAMAS and modified Global Quality Score (mGQS) and content via Total Concent Evaluation (TCE).
    RESULTS: There was high interobserver agreement for DISCERN (intraclass correlation coefficient [ICC]: 0.975; 95% confidence interval [CI]: 0.957-0.985; P: 0.000; P < 0.05) and mGQS (ICC: 0.904; 95% CI: 0.841-0.943; P: 0.000; P < 0.05). Academic source videos had higher DISCERN, GQS, and TCE, revealing both reliability and quality. Also, positive relationship of VPI with mGQS (30.1%) (P: 0.035) and DISCERN (38.1%) (P: 0.007) is detected. The scores revealed 51.9% relationship between mGQS and DISCERN (P: 0.001); and educational quality predictor scores revealed 62.5% relationship between TCE and GQS (P: 0.000).
    CONCLUSION: Despite the limited number of relevant videos, YouTube involves reliable and quality videos that can be used by dentists about learning AIDR.
    Keywords:  YouTube; artificial intelligence; deep learning; dental radiology; internet
  10. BJU Int. 2020 Aug 17.
    Huang MM, Winoker JS, Allaf ME, Matlaga BR, Koo K.
      OBJECTIVE: To assess the quality and accuracy of online videos about the medical management of nephrolithiasis.MATERIALS AND METHODS: To evaluate trends in online interest, we first examined the frequency of worldwide YouTube searches for "kidney stones" from 2015 to 2020. We then queried YouTube with terms related to symptoms and treatment of kidney stones and analyzed English-language videos with over 5,000 views. Quality was assessed using the validated DISCERN instrument. Evidence-based content analysis of video content and viewer comments was performed.
    RESULTS: Online searches for videos about kidney stones doubled between 2015 and 2019 (P<0.001). We analyzed 102 videos with a median of 46,539 views (range 5024-3,631,322). The mean DISCERN score was 3.0 (SD 1.4) out of 5, indicating "moderate" quality; scores were significantly higher for 21 videos (21%) authored by academic hospitals (mean 3.7 vs. 2.8, p=0.02). Inaccurate or non-evidence-based claims were identified in 23 videos (23%); none of the videos authored by academic institutions contained inaccurate claims. Videos with inaccurate statements had more than double the viewer engagement (viewer-generated comments, "thumbs up" and "thumbs down" ratings) compared to videos without inaccuracies (p<0.001). Among viewer comments, 43 videos (43%) included comments with inaccurate or non-evidence-based claims, and a large majority (82 videos, 80%) had "chatbot" recommendations.
    CONCLUSIONS: Interest in YouTube videos about nephrolithiasis has doubled since 2015. While highly-viewed videos vary widely in quality and accuracy, videos produced by academic hospitals have significantly fewer inaccurate claims. Given the high prevalence of stone disease and poor-quality videos, patients should be directed to evidence-based content online.
    Keywords:  Consumer Information; Kidney Stone Disease; Medical Management; Nephrolithiasis; Videos
  11. J Stroke Cerebrovasc Dis. 2020 Sep;pii: S1052-3057(20)30483-3. [Epub ahead of print]29(9): 105065
    Szmuda T, Alkhater A, Albrahim M, Alquraya E, Ali S, Dunquwah RA, Słoniewski P.
      BACKGROUND: Stroke is the second leading cause of death worldwide following ischemic heart disease, and the fifth in the United States. The video-sharing database, YouTube, is the second most popular visited website with more than 2 billion users, thus it's increasingly being used as a medium for delivering health information.AIM: We aimed to evaluate the quality, reliability and audience engagement of stroke-related YouTube videos.
    METHODS: In October 2019 we conducted a search on YouTube using 5 keywords: stroke, brain attack, hemorrhagic stroke, ischemic stroke and transient ischemic attack. We selected the first 30 videos from each search query for further analysis. The validated DISCERN instrument was used (a score of 0-5 per question) to assess the videos by four independent raters. We then recorded qualitative data and quantitative data for each video.
    RESULTS: After sorting through 150 stroke videos, a total of 101 unique YouTube videos met our inclusion criteria. We found that the mean overall quality of YouTube videos according to DISCERN is of fair quality. Most videos (65.3%) were uploaded by hospitals, mentioned the symptoms of stroke (66.3%), had a doctor speaking (60.4%) and contained diagrams (20.8%).
    CONCLUSION: YouTube is a useful source of gathering information about treatment choices for patients and their families as the quality of YouTube videos is fair. The audience engagement suggestions in our paper may help content creators improve the appeal of YouTube videos.
    Keywords:  Internet; Neurology; Neurosurgery; Online; Stroke; YouTube