CoChat

Evaluation of a prototype machine learning tool to semi-automate data extraction for systematic literature reviews

Open access
Panayi, Antonia, Ward, Katherine, Benhadji-Schaff, Amir, Ibanez-Lopez, A Santiago, Xia, Andrew, Barzilay, Regina · 2023 · Systematic Reviews

The DOI matches its Crossref record; 36 of the 36 references it cites were found in a registry or an index.

Abstract

Background: Evidence-based medicine requires synthesis of research through rigorous and time-intensive systematic literature reviews (SLRs), with significant resource expenditure for data extraction from scientific publications. Machine learning may enable the timely completion of SLRs and reduce errors by automating data identification and extraction. Methods: We evaluated the use of machine learning to extract data from publications related to SLRs in oncology (SLR 1) and Fabry disease (SLR 2). SLR 1 predominantly contained interventional studies and SLR 2 observational studies. Predefined key terms and data were manually annotated to train and test bidirectional encoder representations from transformers (BERT) and bidirectional long-short-term memory machine learning models. Using human annotation as a reference, we assessed the ability of the models to identify biomedical terms of interest (entities) and their relations. We also pretrained BERT on a corpus of 100,000 open access clinical publications and/or enhanced context-dependent entity classification with a conditional random field (CRF) model. Performance was measured using the F 1 score, a metric that combines precision and recall. We defined successful matches as partial overlap of entities of the same type. Results: For entity recognition, the pretrained BERT+CRF model had the best performance, with an F 1 score of 73% in SLR 1 and 70% in SLR 2. Entity types identified with the highest accuracy were metrics for progression-free survival (SLR 1, F 1 score 88%) or for patient age (SLR 2, F 1 score 82%). Treatment arm dosage was identified less successfully (F 1 scores 60% [SLR 1] and 49% [SLR 2]). The best-performing model for relation extraction, pretrained BERT relation classification, exhibited F 1 scores higher than 90% in cases with at least 80 relation examples for a pair of related entity types. Conclusions: The performance of BERT is enhanced by pretraining with biomedical literature and by combining with a CRF model. With refinement, machine learning may assist with manual data extraction for SLRs.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1186/s13643-023-02351-w

Cites

36 of 36 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Our development process for refining language models to perform entity recognition and relation extraction. BERT bidirectional encoder representations from transformers, BiLSTM bidirectional long-short-term memory, CRF conditional random field, SLR, systematic literature review
Fig. 1 Our development process for refining language models to perform entity recognition and relation extraction. BERT bidirectional encoder representations from transformers, BiLSTM bidirectional long-short-term memory, CRF conditional random field, SLR, systematic literature reviewCC BY · PubMed Central
Fig. 2 Performance of the pretrained BERT+CRF model across entity types. Panel A presents the relaxed F1 scores and panel B compares actual and predicted entity labels using confusion matrices. In B, some lines do not sum to 100% owing to rounding. BERT bidirectional encoder representations from transformers, CRF conditional random field, eGFR estimated glomerular filtration rate, PFS progression-free survival SLR systematic literature review
Fig. 2 Performance of the pretrained BERT+CRF model across entity types. Panel A presents the relaxed F1 scores and panel B compares actual and predicted entity labels using confusion matrices. In B, some lines do not sum to 100% owing to rounding. BERT bidirectional encoder representations from transformers, CRF conditional random field, eGFR estimated glomerular filtration rate, PFS progression-free survival SLR systematic literature reviewCC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 36 references

Research works printed but not linked: 3

  1. Li T, Higgins JPT, Deeks JJ. Chapter 5. Collecting data. In: Higgins JPT, et al. (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6․3 (updated February 2022). Cochrane, 2022. Available from: https://training.cochrane.org/handbook/current/chapter-05.
  2. Stenetorp P, Pyysalo S, Topić G, et al. brat: a web-based tool for NLP-assisted text annotation. Proceedings of the Demonstrations at the 13th Conference of the European Chapter of the Association for Computational Linguistics; April 2012, Avignon, France. Stroudsburg: Association for Computational Linguistics; 2012. p. 102–7.
  3. Segura-Bedmar I, Martínez Fernández P, Herrero Zazo M. SemEval-2013 Task 9: extraction of drug-drug interactions from biomedical texts (DDIExtraction 2013). In: Second Joint Conference on Lexical and Computational Semantics (*SEM), Volume 2: Proceedings of the Seventh International Workshop on Semantic Evaluation (SemEval 2013); June 2013, Atlanta, GA, USA. Stroudsburg: Association for Computational Linguistics; 2013. p. 341–50.

Cited by

See all 25

Metadata

Authors
Panayi, A., Ward, K., Benhadji-Schaff, A., Ibanez-Lopez, A. S., Xia, A., & Barzilay, R.
Affiliations
  • Takeda (Switzerland) (CH)
  • Massachusetts Institute of Technology (US)
Corresponding author
Panayi, Antonia
Venue
Systematic Reviews
Published
2023-10-06
Volume
12
Issue
1
Pages
187-187
ISSN
2046-4053 (ISSN-L)
OpenAlex
W4387404938
Cited by
23
References
28
Citations by year
2026: 9 · 2025: 11 · 2024: 3
FWCI
0.61
Type
journal article
Language
en
Open access
Licence
Locations
  • Systematic Reviews (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DSpace@MIT (Massachusetts Institute of Technology) (submittedVersion, cc-by-nc)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Takeda Pharmaceuticals International
Topics
  • Health Sciences › Medicine › Health Informatics › Biomedical Text Mining and Ontologies
  • Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
  • Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
Keywords
automated data extraction, systematic reviews, BERT, bidirectional LSTM, entity recognition, relation extraction, conditional random fields, biomedical text mining, Fabry disease, pre-trained language models, clinical trial data
MeSH terms
Machine Learning, Health Expenditures, Humans, Medical Oncology, Evidence-Based Medicine

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Evaluation of a prototype machine learning tool to semi-automate data extraction for systematic literature reviews. https://app.cochat.ai/papers/10.1186/s13643-023-02351-w
The paper
Panayi, A., Ward, K., Benhadji-Schaff, A., Ibanez-Lopez, A. S., Xia, A., & Barzilay, R. (2023). Evaluation of a prototype machine learning tool to semi-automate data extraction for systematic literature reviews. Systematic Reviews. https://doi.org/10.1186/s13643-023-02351-w

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.