bims-librar Biomed News
on Biomedical librarianship
Issue of 2020‒06‒07
eighteen papers selected by
Thomas Krichel
Open Library Society


  1. JMIR Med Inform. 2020 Jun 04. 8(6): e12799
    Massonnaud CR, Kerdelhué G, Grosjean J, Lelong R, Griffon N, Darmoni SJ.
      BACKGROUND: With the continuous expansion of available biomedical data, efficient and effective information retrieval has become of utmost importance. Semantic expansion of queries using synonyms may improve information retrieval.OBJECTIVE: The aim of this study was to automatically construct and evaluate expanded PubMed queries of the form "preferred term"[MH] OR "preferred term"[TIAB] OR "synonym 1"[TIAB] OR "synonym 2"[TIAB] OR …, for each of the 28,313 Medical Subject Heading (MeSH) descriptors, by using different semantic expansion strategies. We sought to propose an innovative method that could automatically evaluate these strategies, based on the three main metrics used in information science (precision, recall, and F-measure).
    METHODS: Three semantic expansion strategies were assessed. They differed by the synonyms used to build the queries as follows: MeSH synonyms, Unified Medical Language System (UMLS) mappings, and custom mappings (Catalogue et Index des Sites Médicaux de langue Française [CISMeF]). The precision, recall, and F-measure metrics were automatically computed for the three strategies and for the standard automatic term mapping (ATM) of PubMed. The method to automatically compute the metrics involved computing the number of all relevant citations (A), using National Library of Medicine indexing as the gold standard ("preferred term"[MH]), the number of citations retrieved by the added terms ("synonym 1"[TIAB] OR "synonym 2"[TIAB] OR …) (B), and the number of relevant citations retrieved by the added terms (combining the previous two queries with an "AND" operator) (C). It was possible to programmatically compute the metrics for each strategy using each of the 28,313 MeSH descriptors as a "preferred term," corresponding to 239,724 different queries built and sent to the PubMed application program interface. The four search strategies were ranked and compared for each metric.
    RESULTS: ATM had the worst performance for all three metrics among the four strategies. The MeSH strategy had the best mean precision (51%, SD 23%). The UMLS strategy had the best recall and F-measure (41%, SD 31% and 36%, SD 24%, respectively). CISMeF had the second best recall and F-measure (40%, SD 31% and 35%, SD 24%, respectively). However, considering a cutoff of 5%, CISMeF had better precision than UMLS for 1180 descriptors, better recall for 793 descriptors, and better F-measure for 678 descriptors.
    CONCLUSIONS: This study highlights the importance of using semantic expansion strategies to improve information retrieval. However, the performances of a given strategy, relatively to another, varied greatly depending on the MeSH descriptor. These results confirm there is no ideal search strategy for all descriptors. Different semantic expansions should be used depending on the descriptor and the user's objectives. Thus, we developed an interface that allows users to input a descriptor and then proposes the best semantic expansion to maximize the three main metrics (precision, recall, and F-measure).
    Keywords:  MEDLINE; Medical Subject Headings; PubMed; bibliographic database; information retrieval; literature search; precision; recall; search strategy; thesaurus
    DOI:  https://doi.org/10.2196/12799
  2. Ber Wiss. 2018 Dec;41(4): 383-386
    Krajewski M.
      
    Keywords:  Archiv; Archivpraktiken; Citizen Science; Einstein; Fermat'sche Vermutung; Fermats Last Theorem; Gutachten; Populärwissenschaft; archival practices; archives; review; scientific ecognition; vernacular science; wissenschaftliche Anerkennung
    DOI:  https://doi.org/10.1002/bewi.201801917
  3. BMC Med Inform Decis Mak. 2020 Jun 05. 20(1): 102
    Georgsson S, Carlsson T.
      BACKGROUND: High-quality information is essential if clients who request an abortion are to reach informed decisions and feel prepared for the procedure, but little is known concerning the readability of web-based sources containing such material. The aim was to investigate the readability of web-based information about induced abortion.METHODS: The search engine Google was used to identify web pages about induced abortion, written in the English language. A total of 240 hits were screened and 236 web pages fulfilled the inclusion criteria. After correcting for duplicate hits, 185 web pages were included. The readability of the text-based content of each web page was determined with Flesch Kincaid Grade Level, Gunning Fog Index, Coleman-Liau Index, Simple Measure of Gobbledygook, and Flesch Reading Ease. Data were analyzed with descriptive statistics, Pearson's correlation coefficient and Kruskal-Wallis with Dunn's test as post hoc analysis.
    RESULTS: Across all grade level measures, a small minority of the web pages had a readability corresponding to elementary school (n < 3, 1%), while the majority had readability corresponding to senior high school or above (n > 153, 65%). The means of the grade level measures ranged between 10.5 and 13.1, and the mean Flesch Reading Ease score was 45.3 (SD 13.6). Only weak correlations (rho < 0.2) were found between the readability measures and search rank in the hit lists. Consistently, web pages affiliated with health care had the least difficult readability and those affiliated with scientific sources had the most difficult readability.
    CONCLUSIONS: Overall, web-based information about induced abortions has difficult readability. Incentives are needed to improve the readability of these texts and ensure that clients encounter understandable information so that they may reach informed decisions and feel adequately prepared when requesting an abortion.
    Keywords:  Consumer health information; Induced abortion; Quality; Readability; World wide web
    DOI:  https://doi.org/10.1186/s12911-020-01132-y
  4. J Med Internet Res. 2020 Jun 05. 22(6): e20021
    Wang PW, Lu WH, Ko NY, Chen YL, Li DJ, Chang YP, Yen CF.
      BACKGROUND: People obtain information on the coronavirus disease (COVID-19) from the internet and other sources. Understanding the factors related to such information sources aids health professionals in educating individuals.OBJECTIVE: This study used data collected from the online survey study on COVID-19 in Taiwan to examine what major COVID-19 information sources are available and which sources are significantly related to the self-confidence of people in coping with COVID-19 in Taiwan.
    METHODS: A total of 1904 participants (1270 non-health-care workers and 634 health care workers) were recruited from the Facebook advertisement. Their major sources of information about COVID-19, the relationships between the sources and demographic factors, and the relationships between the sources and the self-confidence in coping with COVID-19 were surveyed.
    RESULTS: Most Taiwanese people relied on the internet for COVID-19 information. Many respondents also used a variety of sources of information on COVID-19; such variety was associated with sex, age, and the level of worry toward COVID-19, as well as if one was a health care worker. For health care workers, the use of formal lessons as an information source was significantly associated with better self-confidence in coping with COVID-19. The significant association between receiving information from more sources and greater self-confidence was found only in health care workers but not in non-health-care workers.
    CONCLUSIONS: Medical professionals should consider subgroups of the population when establishing various means to deliver information on COVID-19.
    Keywords:  COVID-19; Facebook; confidence; coping; information; internet; mental health; online health information; social media; survey
    DOI:  https://doi.org/10.2196/20021
  5. J Pediatr Orthop. 2020 Jul;40(6): 310-313
    Sobel AD, Ramirez JM, Walsh DF, Defroda SF, Cruz AI.
      INTRODUCTION: Given the rapidly increasing population of Spanish-speaking patients in the United States, medical providers must have the capability to effectively communicate both with pediatric patients and their caregivers. The purpose of this study was to query the Spanish language proficiency of pediatric orthopaedic surgeons, assess the educational resources available to Spanish-speaking patients and their families, and identify the barriers to care at academic pediatric orthopaedic centers.METHODS: The Web sites of medical centers within the United States that have pediatric orthopaedic surgery fellowships recognized by the Pediatric Orthopaedic Society of North America (POSNA) were accessed. Web sites were investigated for a health library as well as the availability of interpreter services. Profiles of attending surgeons within each Pediatric Orthopaedic Department were evaluated for evidence of Spanish proficiency as well as educational qualifications. Centers were contacted by phone to determine if the resources and physicians who could converse in Spanish were different than what was readily available online and if automated instructions in Spanish or a person who could converse in Spanish were available.
    RESULTS: Forty-six centers with 44 fellowship programs were identified. The profiles of 12 of 334 (3.6%) surgeons who completed pediatric orthopaedic fellowships indicated Spanish proficiency. Seventeen physicians (5.1%) were identified as proficient in Spanish after phone calls. Thirty-eight pediatric orthopaedic centers (82.6%) noted interpreter service availability online, although services varied from around-the-clock availability of live interpreters to interpreter phones. When contacted by phone, 45 of 46 centers (97.8%) confirmed the availability of any interpreter service for both inpatient and outpatient settings. Sixteen centers (34.8%) had online information on orthopaedic conditions or surgical care translated into Spanish. Twenty centers (43.5%) did not have automated phone messages in Spanish or live operators that spoke Spanish.
    CONCLUSIONS: There is a scarcity of surgical providers in pediatric orthopaedic centers proficient in Spanish, demonstrating a large discrepancy with the growing Hispanic population. Interpreter services are widely available, although there is variability in the services provided. Considerable barriers exist to Spanish-speaking patients who attempt to access care by phone or online.
    DOI:  https://doi.org/10.1097/BPO.0000000000001466
  6. Inform Health Soc Care. 2020 Jun 02. 1-11
    Ahmadian L, Khajouei R, Kamali S, Mirzaee M.
      Pregnant women get information about pregnancy andchild-birth from many sources, including the Internet. There is alack of evidence about the extent to which pregnant women usethis source. This study aimed to investigate the use of the Internetby pregnant women to search for information about pregnancyand childbirth. This study was conducted in Kerman, Iran. Threehundred eighty-five pregnant women waiting for their appointmentswith obstetricians/gynecologists participated in the survey byfilling out a questionnaire. The most common searched topicswere nutrition in pregnancy (81%), fetal development (67%), andcomplications of pregnancy (49%). The most popular sources ofinformation were physicians (61%), the Internet (51%), and printedsources (41%), respectively. More than half of the participantsdid not share the retrieved information from the Internet with theirhealth professionals. After being examined by a physician, 43% ofthe participants searched the Internet about the discussed topic.Beside physicians, the Internet is the second common source ofpregnancy and childbirth information for Iranian pregnant women.Future studies are needed to analyze the quality and accuracy ofonline pregnancy and childbirth information.
    Keywords:  Pregnancy; childbirth; information; internet; online; pregnant woman; world Wide Web
    DOI:  https://doi.org/10.1080/17538157.2020.1769106
  7. BMC Public Health. 2020 Jun 01. 20(1): 832
    Huang J, Zhang S, Xiao Q, Cao Y, Li B.
      BACKGROUND: Candida auris is a novel Candida species, and has emerged globally as a multidrug-resistant health care-associated fungal pathogen. YouTube™ (http://www.youtube.com) as the largest free video-sharing website is increasingly used to search health information. Thus, the aim of this study was to evaluate the content, reliability and quality of YouTube™ videos regarding Candida auris infection, and to identify whether it is a useful resource for people.METHODS: The YouTube™ was used to search systematically for videos using the keywords: "Candida auris infection" and "Candida auris". Strict inclusion and exclusion criteria were used to select the videos. The videos were reviewed and scored by two independent reviewers and recorded the "title", "length", "views", "comments", "dislike", "like", "posted days" and "category of videos". The videos were categorized as "poor", "good" and "excellent" by the score. The DISCERN tool was used to assess the reliability of the YouTube™ videos.
    RESULTS: Seventy-six videos were included in final analysis in our study. Most videos (59.2%, 55/76) had better quality. There were no statistically significant differences between groups in respect of the number of likes, dislikes, views, comments, percentage positivity, likebility, view rate and viewers' interaction. Length and posted days were significantly associated with the classification. The videos were categorized as "educational video", "new report", "personal experience and blog entertainment" and "interview". Significant differences were found in the source of videos and the characteristics of the individuals appearing in a video between the groups.
    CONCLUSION: YouTube™ has striking potential to be an effective user-friendly learning interface for people to obtain information of Candida auris infection.
    Keywords:  Candida auris; Infection; Internet; Reliability; YouTube™
    DOI:  https://doi.org/10.1186/s12889-020-08731-4
  8. Eur J Obstet Gynecol Reprod Biol. 2020 Apr 18. pii: S0301-2115(20)30172-X. [Epub ahead of print]250 224-230
    Hirsch M, Wojtaszewska A, Saridogan E, Mavrelos D, Barker C, Duffy JMN.
      OBJECTIVE: We aim to evaluate the accuracy, quality, and readability of online patient information concerning fibroids.STUDY DESIGN: We searched the most popular Internet search engine: Google.com. We developed a search strategy in consultation with patients with fibroids, to identify relevant websites. Two independent authors screened the search results. Websites were evaluated using validated instruments across three domains, including assessments of: [1] quality (DISCERN instrument; range 0-85); [2] readability (Flesch-Kincaid instrument; range 0-100); and [3] accuracy. Accuracy was assessed using evidence-based statements. We summarised this data narratively including the use of figures and tables.
    RESULTS: We identified 750 websites, of which 48 were included. Over a third of websites did not attribute authorship and almost half the included websites did not report the sources of information or academic references. No website provided written patient information in line with recommendations from the American Medical Association. A minority (18%) of websites were assessed as high quality. Twelve webpages provided only accurate statements. Available information was, in general, skewed towards the surgical management of fibroids. No website scored highly across all three domains.
    CONCLUSION: In the unlikely event that a website reports high quality and accurate health information, it is typically challenging for a lay audience to comprehend. Healthcare professionals and the wider community, should inform women with fibroids of the risk of outdated, inaccurate, or even dangerous information online. The implementation of an Information Standard certification will incentivise providers of online information to establish and adhere to codes of conduct. VIDEO ABSTRACT.
    Keywords:  Accuracy; Fibroids; Online information; Patients; Quality; Readability
    DOI:  https://doi.org/10.1016/j.ejogrb.2020.04.004
  9. Ophthalmol Retina. 2020 Jun 01. pii: S2468-6530(20)30215-3. [Epub ahead of print]
    Rayess N, Li AS, Do DV, Rahimy E.
      PURPOSE: Patients increasingly use the internet to access health-related information to further understand their treatments and conditions. This study compares the quality, accountability, readability, accessibility and presence of translation between private and academic online source material available to the public regarding intravitreal injections.DESIGN: Cross-sectional analysis PARTICIPANTS: Top 20 websites on a Google search for the terms 'eye injections', 'intravitreal injections' and 'anti-VEGF injections'.
    METHODS: Websites were classified as private or academic. Quality and accountability were assessed using the internationally recognized DISCERN criteria and the Health on the Net (HONcode). All 20 sites were independently graded by 2 retinal physicians and differences were adjudicated by a third experienced retinal physician. Readability was evaluated using an online tool that provides a consensus readability grade level. The presence of and languages available for translation were recorded. The top 5 ranked websites' content quality, accountability and readability was also compared with the other 15 websites.
    MAIN OUTCOME MEASURES: Primary outcome measure is comparing the DISCERN and HONcode quality and accountability scores between academic and private websites. Secondary outcome measures include evaluating readability, accessibility and presence of translation (in particular, Spanish).
    RESULTS: Eleven academic and 9 private websites were included. The overall mean score using DISCERN criteria for the academic websites (3.11±0.46) was significantly higher than that of private websites (2.23±0.61; p<0.007). Similarly, out of a possible total 14 points for the HONcode, the average quality score for academic websites (10.91±2.66) was higher compared to private websites (6.44±3.36; p<0.009). The mean consensus reading grade level was similar between academic (11.73±1.68) and private websites (11.78±1.48; p=0.94). Spanish translation was offered by only 7 of the 20 websites (5 academic and 2 private websites).
    CONCLUSION: The overall quality and accountability of online content for academic sites was significantly higher compared to private websites. Translation was rarely provided and the readability grade level was significantly higher for both groups than recommended. Improving the quality, accountability, readability, accessibility and incorporating translation in websites can help improve patients' health literacy regarding intravitreal injections, potentially leading to increased adherence to therapy plans and improved treatment outcomes.
    DOI:  https://doi.org/10.1016/j.oret.2020.05.019
  10. J Paediatr Child Health. 2020 Jun 01.
    McKinnon KA, H Y Caldwell P, Scott KM.
      AIM: Adolescents increasingly use smartphones to look up online health information. This pilot study aimed to explore the search and assessment strategies of adolescents looking for online health information.METHODS: We performed an observed, practical task on mobile devices, followed by a semi-structured interview with adolescent patients at The Children's Hospital at Westmead, Sydney. Observational data were analysed using an observation rubric, and interviews were transcribed verbatim. Data were analysed through inductive thematic analysis using line-by-line coding and the constant comparative process.
    RESULTS: The research was undertaken with 10 participants. Three themes were identified: (i) participants' searching strategies to find online health information; (ii) techniques for assessing relevance; and (iii) techniques for assessing credibility. These themes demonstrated that most participants accessed online health information due to its ease and accessibility but failed to assess credibility. Most prioritised relevance of information over credibility, determined by their personal knowledge and experience. Our results indicate that there was a large discrepancy between adolescents' ability to search for and assess online health information and their perceived ability. This demonstrates a discrepancy between perceived and performance-based eHealth literacy and highlights poor critical self-awareness, which can prevent adolescents from seeking help. This may underlie the biggest challenge in adolescents' access of online health information and highlights the need for education.
    CONCLUSIONS: Many adolescents' search and appraisal ability is negatively impacted by low eHealth literacy. These adolescents' inability to recognise their need for assistance in improving their search and assessment strategies highlights the need for multi-stage education.
    Keywords:  adolescent; digital health literacy; eHealth literacy; health promotion; information seeking behaviour
    DOI:  https://doi.org/10.1111/jpc.14918
  11. BMC Public Health. 2020 Jun 05. 20(1): 860
    Litzkendorf S, Frank M, Babac A, Rosenfeldt D, Schauer F, Hartz T, Graf von der Schulenburg JM.
      BACKGROUND: Finding reliable information on one of more than 7000 rare diseases is a major challenge for those affected. Since rare diseases are defined only by the prevalence criterion, a multitude of heterogeneous diseases are included. Common to all, however, are difficulties regarding information access. Even though various quantitative studies have analyzed the use of different information sources for specific rare diseases, little is known about the use of information sources for different rare diseases, how users rate these information sources based on their experiences, and how the use and importance of these information sources change over time.METHODS: Fifty-five patients with a variety of rare diseases and 13 close relatives participated in qualitative interviews. For these interviews, a semi-structured guideline was developed, piloted, and revised. Data analysis involved a qualitative content analysis developed by Philipp Mayring.
    RESULTS: The participants considered internet as the most important and widespread information source, especially for early information. Although patients have difficulty dealing with information obtained online, they consider online searching a quick and practical option to gather information. During the course of the disease, personal contact partners, especially self-help associations and specialized doctors, become more important. This is also because information provided online is sometimes insufficiently detailed to answer their information needs, which can be complemented by information from doctors and self-help.
    CONCLUSIONS: People rarely use just one type of source, but rather refer to different sources and informants. The source used depends on the type of information sought as well as other person-related factors such as preexisting knowledge and the disease stage. To improve people's information searching and connect them with medical specialists in rare diseases, a central information portal on rare diseases might be a suitable access point to provide free and quality assured information for patients, caregivers, and physicians. This would allow not only patients but also doctors to find quality assured information on symptoms and therapies as well as patient associations and specialized doctors.
    Keywords:  Content analysis; Health information seeking; Informants; Information sources; Online information; Qualitative research; Rare diseases; Self-help; Written information
    DOI:  https://doi.org/10.1186/s12889-020-08926-9
  12. AMIA Jt Summits Transl Sci Proc. 2020 ;2020 579-588
    Shenoi SJ, Ly V, Soni S, Roberts K.
      Precision medicine focuses on developing new treatments based on an individual's genetic, environmental, and lifestyle profile. While this data-driven approach has led to significant advances, retrieving information specific to a patient's condition has proved challenging for oncologists due to the large volume of data. In this paper, we propose the PRecIsion Medicine Robust Oncology Search Engine (PRIMROSE) for cancer patients that retrieves scientific articles and clinical trials based on a patient's condition, genetic profile, age, and gender. Our search engine utilizes Elasticsearch indexes for information storage and retrieval, and we developed a knowledge graph for query expansion in order to improve recall. Additionally, we experimented with machine learning and learning-to-rank components to the search engine and compared the results of the two approaches. Finally, we developed a front-facing ReactJS website and a REST API for connecting with our search engine. The development of this front-facing website allows for easy access to our system by healthcare providers.
  13. AMIA Jt Summits Transl Sci Proc. 2020 ;2020 53-62
    Anani M, Kuntz M, Kahanda I.
      Research Domain Criteria (RDoC), which is a recently introduced framework for mental illness, utilizes various units of analysis from genetics, neural circuits, etc., for accurate multi-dimensional classification of mental illnesses. Due to the large amount of relevant biomedical research available, automating the process of extracting evidence from the literature to assist with the curation of the RDoC matrix is essential for processing the full breadth of data in an accurate and cost-effective manner. In this work, we formulate the task of information retrieval of brain research literature from general PubMed abstracts. We develop BRret (Brain Research retriever), a novel algorithm for brain research related article retrieval. We use a large dataset of PubMed abstracts annotated with RDoC concepts to demonstrate the effectiveness of BRret. To the best of our knowledge, this is the first study aimed at automated retrieval of brain research related literature.
  14. AMIA Jt Summits Transl Sci Proc. 2020 ;2020 561-568
    Savery ME, Rogers WJ, Pillai M, Mork JG, Demner-Fushman D.
      Chemical entity recognition is essential for indexing scientific literature in the MEDLINE database at the National Library of Medicine. However, the tool currently used to suggest terms for indexing, the Medical Text Indexer, was not originally conceived as a chemical recognition tool. It has instead been adapted to the task via its use of MetaMap and the addition of in-house patterns and rules. In order to develop a tool more suitable for chemical recognition, we have created a collection of 200 MEDLINE titles and abstracts annotated with genes, proteins, inorganic and organic chemicals, as well as other biological molecules. We use this collection to evaluate eleven chemical entity recognition systems, where we seek to identify a tool that effectively recognizes chemical entities for indexing and also performs well on chemical recognition beyond the indexing task. We observe the highest performance with a SciBERT ensemble.
  15. BMC Med Res Methodol. 2020 Jun 01. 20(1): 138
    Goossen K, Hess S, Lunny C, Pieper D.
      BACKGROUND: When conducting an Overviews of Reviews on health-related topics, it is unclear which combination of bibliographic databases authors should use for searching for SRs. Our goal was to determine which databases included the most systematic reviews and identify an optimal database combination for searching systematic reviews.METHODS: A set of 86 Overviews of Reviews with 1219 included systematic reviews was extracted from a previous study. Inclusion of the systematic reviews was assessed in MEDLINE, CINAHL, Embase, Epistemonikos, PsycINFO, and TRIP. The mean inclusion rate (% of included systematic reviews) and corresponding 95% confidence interval were calculated for each database individually, as well as for combinations of MEDLINE with each other database and reference checking.
    RESULTS: Inclusion of systematic reviews was higher in MEDLINE than in any other single database (mean inclusion rate 89.7%; 95% confidence interval [89.0-90.3%]). Combined with reference checking, this value increased to 93.7% [93.2-94.2%]. The best combination of two databases plus reference checking consisted of MEDLINE and Epistemonikos (99.2% [99.0-99.3%]). Stratification by Health Technology Assessment reports (97.7% [96.5-98.9%]) vs. Cochrane Overviews (100.0%) vs. non-Cochrane Overviews (99.3% [99.1-99.4%]) showed that inclusion was only slightly lower for Health Technology Assessment reports. However, MEDLINE, Epistemonikos, and reference checking remained the best combination. Among the 10/1219 systematic reviews not identified by this combination, five were published as websites rather than journals, two were included in CINAHL and Embase, and one was included in the database ERIC.
    CONCLUSIONS: MEDLINE and Epistemonikos, complemented by reference checking of included studies, is the best database combination to identify systematic reviews on health-related topics.
    Keywords:  Databases; Overview of reviews; Review methods; Search strategy; Systematic reviews; Umbrella review
    DOI:  https://doi.org/10.1186/s12874-020-00983-3
  16. AMIA Jt Summits Transl Sci Proc. 2020 ;2020 288-297
    Kamdar MR, Stanley CE, Carroll M, Wogulis L, Dowling W, Deus HF, Samarasinghe M.
      Knowledge graphs have been shown to significantly improve search results. Usually populated by subject matter experts, relations therein need to keep up to date with medical literature in order for search to remain relevant. Dynamically identifying text snippets in literature that confirm or deny knowledge graph triples is increasingly becoming the differentiator between trusted and untrusted medical decision support systems. This work describes our approach to mapping triples to medical text. A medical knowledge graph is used as a source of triples that are used to find matching sentences in reference text. Our unsupervised approach uses phrase embeddings and cosine similarity measures, and boosts candidate text snippets when certain key concepts exist. Using this approach, we can accurately map semantic relations within the medical knowledge graph to text snippets with a precision of 61.4% and recall of 86.3%. This method will be used to develop a novel application in the future to retrieve medical relations and corroborating snippets from medical text given a user query.