A semi-supervised approach using label propagation to support citation screening

Kontonatsios, Georgios; Brockmeier, Austin J.; Przybyła, Piotr; McNaught, John; Mu, Tingting; Goulermas, John Y.; Ananiadou, Sophia

doi:10.1016/j.jbi.2017.06.018

Cited by 37 publications

(53 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although a reviewer can use the system's relevance scores in many ways, relevancy‐based prioritisation, where the references the system deems as the most likely to be relevant are screened, is the suggested approach. Prioritising the relevant references earlier in the screening process is considered beneficial for gaining an understanding of the area or in cases when the reviewer does not plan to screen the whole collection …”

Section: Methodsmentioning

confidence: 99%

“…Subsequent studies have explored different feature spaces (words or multiword patterns, MeSH terms from PubMed metadata, unified medical language system (UMLS) for nomenclature, and topical or thematic features) and machine learning models and techniques such as naive Bayes, support vector machine (SVM), and logistic regression. Others have incorporated out‐of‐topic inclusions and unscreened references …”

Section: Related Workmentioning

confidence: 99%

“…Others have incorporated out-of-topic inclusions 53 and unscreened references. 30,31 Most previous studies of machine learning for systematic reviews have a fixed training and test set of references. In these cases, a user screens a portion of references (either random or based on publication year for an update); the machine learns from the screened portion and predicts the relevant references within the remainder of references (the test set).…”

Section: Related Workmentioning

confidence: 99%

“…The potential benefit of incorporating machine learning into the systematic review toolkit has been showcased by numerous retrospective studies (simulations of the screening process using previously screened collections 9,13,[15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31] and some prospective studies 32,33 that have explored different machine learning approaches. Yet, before semi-automated tools with functionality like RobotAnalyst become widely adopted within the systematic review community, there is a need for real-world evaluation, user feedback, and discussion to understand the benefits and potential risks.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Prioritising references for systematic reviews with RobotAnalyst: A user study

Przybyła

Brockmeier

Kontonatsios

et al. 2018

Research Synthesis Methods

Self Cite

100

View full text Add to dashboard Cite

Screening references is a time‐consuming step necessary for systematic reviews and guideline development. Previous studies have shown that human effort can be reduced by using machine learning software to prioritise large reference collections such that most of the relevant references are identified before screening is completed. We describe and evaluate RobotAnalyst, a Web‐based software system that combines text‐mining and machine learning algorithms for organising references by their content and actively prioritising them based on a relevancy classification model trained and updated throughout the process. We report an evaluation over 22 reference collections (most are related to public health topics) screened using RobotAnalyst with a total of 43 610 abstract‐level decisions. The number of references that needed to be screened to identify 95% of the abstract‐level inclusions for the evidence review was reduced on 19 of the 22 collections. Significant gains over random sampling were achieved for all reviews conducted with active prioritisation, as compared with only two of five when prioritisation was not used. RobotAnalyst's descriptive clustering and topic modelling functionalities were also evaluated by public health analysts. Descriptive clustering provided more coherent organisation than topic modelling, and the content of the clusters was apparent to the users across a varying number of clusters. This is the first large‐scale study using technology‐assisted screening to perform new reviews, and the positive results provide empirical evidence that RobotAnalyst can accelerate the identification of relevant studies. The results also highlight the issue of user complacency and the need for a stopping criterion to realise the work savings.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Prioritising references for systematic reviews with RobotAnalyst: A user study

Przybyła

Brockmeier

Kontonatsios

et al. 2018

Research Synthesis Methods

Self Cite

100

View full text Add to dashboard Cite

show abstract

“…Using the prioritization feature of RobotAnalyst, screening work was reduced by 88% in the present review. 7,8 Inclusion criteria for the studies were comparison of biosimilar and originator insulins in adult patients (aged ≥18 years) with type Weighted mean differences and pooled odds ratios (ORs) were calculated by DerSimonian-Laird random-effects meta-analysis, 10 with 0.5 being added for correction if no events were reported in a study treatment group. Documented hypoglycaemia was defined as plasma glucose ≤3.9 mmol/L (≤70 mg/dL) and severe hypoglycaemia as an episode requiring assistance.…”

Section: Methodsmentioning

confidence: 99%

Biosimilar vs originator insulins: Systematic review and meta‐analysis

Yamada

Kamata

Ishinohachi

et al. 2018

Diabetes Obesity Metabolism

Self Cite

View full text Add to dashboard Cite

Biosimilar insulins have expanded the treatment options for diabetes. We compared the clinical efficacy and safety of biosimilar insulins with those of originator insulins by conducting a meta-analysis. A random-effects meta-analysis was performed on randomized controlled trials comparing biosimilar and originator insulins in adults with diabetes. Studies were obtained by searching electronic databases up to December 2017. Ten trials, in a total of 4935 patients, were assessed (2 trials each on LY2963016, MK-1293, Mylan's insulin glargine and SAR342434, and 1 trial each on FFP-112 and Basalog). The meta-analysis found no differences between long-acting biosimilar and originator insulins with regard to reduction in glycated haemoglobin at 24 weeks (0.04%, 95% confidence interval [CI] -0.01, 0.08; P for efficacy = .14, I = 0%) or at 52 weeks (0.03%, 95% CI -0.04, 0.1), or reduction in fasting plasma glucose (0.08 mmol/L, 95% CI 0.36, 0.53), hypoglycaemia (odds ratio 0.99, 95% CI 0.96, 1.03), mortality, injection site reactions, insulin antibodies and allergic reactions. Analyses stratified by type of diabetes and prior insulin use yielded similar findings. Similarly, no significant differences were found between short-acting biosimilar and originator insulins. In summary, our meta-analysis showed no significant differences in clinical efficacy and safety, including immune reactions, between biosimilar and originator insulins. Biosimilar insulins can increase access to modern insulin therapy and reduce medical costs.

show abstract