Literature DB >> 34039433

Text mining to support abstract screening for knowledge syntheses: a semi-automated workflow.

Ba' Pham1, Jelena Jovanovic2, Ebrahim Bagheri3, Jesmin Antony1, Huda Ashoor1, Tam T Nguyen3, Patricia Rios1, Reid Robson1, Sonia M Thomas1, Jennifer Watt1, Sharon E Straus1, Andrea C Tricco4,5,6.   

Abstract

BACKGROUND: Current text mining tools supporting abstract screening in systematic reviews are not widely used, in part because they lack sensitivity and precision. We set out to develop an accessible, semi-automated "workflow" to conduct abstract screening for systematic reviews and other knowledge synthesis methods.
METHODS: We adopt widely recommended text-mining and machine-learning methods to (1) process title-abstracts into numerical training data; and (2) train a classification model to predict eligible abstracts. The predicted abstracts are screened by human reviewers for ("true") eligibility, and the newly eligible abstracts are used to identify similar abstracts, using near-neighbor methods, which are also screened. These abstracts, as well as their eligibility results, are used to update the classification model, and the above steps are iterated until no new eligible abstracts are identified. The workflow was implemented in R and evaluated using a systematic review of insulin formulations for type-1 diabetes (14,314 abstracts) and a scoping review of knowledge-synthesis methods (17,200 abstracts). Workflow performance was evaluated against the recommended practice of screening abstracts by 2 reviewers, independently. Standard measures were examined: sensitivity (inclusion of all truly eligible abstracts), specificity (exclusion of all truly ineligible abstracts), precision (inclusion of all truly eligible abstracts among all abstracts screened as eligible), F1-score (harmonic average of sensitivity and precision), and accuracy (correctly predicted eligible or ineligible abstracts). Workload reduction was measured as the hours the workflow saved, given only a subset of abstracts needed human screening.
RESULTS: With respect to the systematic and scoping reviews respectively, the workflow attained 88%/89% sensitivity, 99%/99% specificity, 71%/72% precision, an F1-score of 79%/79%, 98%/97% accuracy, 63%/55% workload reduction, with 12%/11% fewer abstracts for full-text retrieval and screening, and 0%/1.5% missed studies in the completed reviews.
CONCLUSION: The workflow was a sensitive, precise, and efficient alternative to the recommended practice of screening abstracts with 2 reviewers. All eligible studies were identified in the first case, while 6 studies (1.5%) were missed in the second that would likely not impact the review's conclusions. We have described the workflow in language accessible to reviewers with limited exposure to natural language processing and machine learning, and have made the code available to reviewers.

Entities:  

Keywords:  Abstract screening; Automation; Classification model; Machine learning; Natural language processing; Scoping review; Systematic review; Text mining

Year:  2021        PMID: 34039433     DOI: 10.1186/s13643-021-01700-x

Source DB:  PubMed          Journal:  Syst Rev        ISSN: 2046-4053


  15 in total

1.  Estimating time to conduct a meta-analysis from number of citations retrieved.

Authors:  I E Allen; I Olkin
Journal:  JAMA       Date:  1999-08-18       Impact factor: 56.272

Review 2.  A survey of current work in biomedical text mining.

Authors:  Aaron M Cohen; William R Hersh
Journal:  Brief Bioinform       Date:  2005-03       Impact factor: 11.622

Review 3.  Deep learning.

Authors:  Yann LeCun; Yoshua Bengio; Geoffrey Hinton
Journal:  Nature       Date:  2015-05-28       Impact factor: 49.962

4.  RysannMD: A biomedical semantic annotator balancing speed and accuracy.

Authors:  John Cuzzola; Jelena Jovanović; Ebrahim Bagheri
Journal:  J Biomed Inform       Date:  2017-05-26       Impact factor: 6.317

5.  Improving the conduct of systematic reviews: a process mining perspective.

Authors:  Ba' Pham; Ebrahim Bagheri; Patricia Rios; Asef Pourmasoumi; Reid C Robson; Jeremiah Hwee; Wanrudee Isaranuwatchai; Nazia Darvesh; Matthew J Page; Andrea C Tricco
Journal:  J Clin Epidemiol       Date:  2018-07-02       Impact factor: 6.437

Review 6.  Knowledge synthesis methods for generating or refining theory: a scoping review reveals that little guidance is available.

Authors:  Andrea C Tricco; Jesmin Antony; Charlene Soobiah; Monika Kastner; Elise Cogo; Heather MacDonald; Jennifer D'Souza; Wing Hui; Sharon E Straus
Journal:  J Clin Epidemiol       Date:  2016-02-15       Impact factor: 6.437

Review 7.  Use of cost-effectiveness analysis to compare the efficiency of study identification methods in systematic reviews.

Authors:  Ian Shemilt; Nada Khan; Sophie Park; James Thomas
Journal:  Syst Rev       Date:  2016-08-17

8.  Rayyan-a web and mobile app for systematic reviews.

Authors:  Mourad Ouzzani; Hossam Hammady; Zbys Fedorowicz; Ahmed Elmagarmid
Journal:  Syst Rev       Date:  2016-12-05

9.  Moving toward the automation of the systematic review process: a summary of discussions at the second meeting of International Collaboration for the Automation of Systematic Reviews (ICASR).

Authors:  Annette M O'Connor; Guy Tsafnat; Stephen B Gilbert; Kristina A Thayer; Mary S Wolfe
Journal:  Syst Rev       Date:  2018-01-09

10.  Epidemiology and Reporting Characteristics of Systematic Reviews of Biomedical Research: A Cross-Sectional Study.

Authors:  Matthew J Page; Larissa Shamseer; Douglas G Altman; Jennifer Tetzlaff; Margaret Sampson; Andrea C Tricco; Ferrán Catalá-López; Lun Li; Emma K Reid; Rafael Sarkis-Onofre; David Moher
Journal:  PLoS Med       Date:  2016-05-24       Impact factor: 11.069

View more
  3 in total

1.  Data extraction methods for systematic review (semi)automation: A living systematic review.

Authors:  Lena Schmidt; Babatunde K Olorisade; Luke A McGuinness; James Thomas; Julian P T Higgins
Journal:  F1000Res       Date:  2021-05-19

Review 2.  Past and future uses of text mining in ecology and evolution.

Authors:  Maxwell J Farrell; Liam Brierley; Anna Willoughby; Andrew Yates; Nicole Mideo
Journal:  Proc Biol Sci       Date:  2022-05-18       Impact factor: 5.530

3.  Automating risk of bias assessment in systematic reviews: a real-time mixed methods comparison of human researchers to a machine learning system.

Authors:  Patricia Sofia Jacobsen Jardim; Christopher James Rose; Heather Melanie Ames; Jose Francisco Meneses Echavez; Stijn Van de Velde; Ashley Elizabeth Muller
Journal:  BMC Med Res Methodol       Date:  2022-06-08       Impact factor: 4.612

  3 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.