Meta-analysis and systematic reviews

Onlinemeta: A Web Server for Meta-Analysis Based on Shiny Framework

Onlinemeta provides a web-based tool for conducting meta-analysis and network meta-analysis with various analytical modules and visualizations.

Reporting quality, effect sizes, and biases for aging interventions: a methodological appraisal of the DrugAge database

The paper evaluates the DrugAge database, finding poor reporting quality and inconsistent effect sizes in anti-aging preclinical studies across species.

RAPID: Reliable and efficient Automatic generation of submission rePortIng checklists with large language moDels

RAPID is an automated tool using large language models to generate medical reporting checklists, achieving high accuracy on CONSORT and CONSORT-AI datasets.

Bayesian p-curve Mixture Models as a Tool to Dissociate Effect Size and Effect Prevalence

This paper introduces a Bayesian method to estimate effect prevalence and size by clustering participant data, outperforming existing methods when effect size is uncertain.

Re-expressing coefficients from regression models for inclusion in a meta-analysis

The paper evaluates methods to re-express regression coefficients for meta-analysis when independent variables are log-transformed or not, finding that all methods introduce bias and recommending stratification instead.

The Automated Systematic Search Deduplicator (ASySD): a rapid, open-source, interoperable tool to remove duplicate citations in biomedical systematic reviews.

ASySD is a tool that efficiently removes duplicate citations in biomedical systematic reviews with high sensitivity and reliability.

A consensus score to combine inferences from multiple centres

The paper introduces a consensus score method to combine statistical results from multiple resources and applies it to detect sexual dimorphism in haematological data from the IMPC.

Risk of Bias Assessment in Preclinical Literature using Natural Language Processing

This paper applies natural language processing to assess risk of bias in preclinical literature, achieving high F1 scores for several bias items using neural models and BERT.

Estimating the prevalence of missing experiments in a neuroimaging meta-analysis

The paper estimates the prevalence of missing non-significant experiments in a neuroimaging meta-analysis using a zero-truncated model applied to BrainMap data.

Sample size evolution in neuroimaging research: an evaluation of highly-cited studies (1990-2012) and of latest practices (2017-2018) in high-impact journals

The paper evaluates sample size trends in neuroimaging studies from 1990-2012 and 2017-2018, finding increasing sample sizes and low use of pre-study power calculations.

Revisiting and expanding the meta-analysis of variation: The log coefficient of variation ratio, lnCVR

This paper reviews and proposes new estimators for effect size statistics lnVR and lnCVR to compare variability between two groups in meta-analyses, emphasizing the importance of accounting for group dependence.

Machine learning algorithms for systematic reviews: reducing workload in a preclinical review of animal studies and reducing human screening error

This paper applies machine learning to citation screening in systematic reviews of animal studies, improving sensitivity and identifying human errors.

Effect size and statistical power in the rodent fear conditioning literature - a systematic review

This study reviews effect sizes and statistical power in rodent fear conditioning research, finding low power and inconsistent reporting of these metrics.