Analysis of article screening and data extraction performance by an AI systematic literature review platform
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Artificial intelligence (AI) tools can automate systematic literature reviews (SLRs), saving time and resources. An AI platform, ISLaR 2.0, showed high accuracy in screening and data extraction, significantly reducing review times.
Area Of Science
- Health Research Methodology
- Artificial Intelligence in Healthcare
- Systematic Literature Review Automation
Background
- Systematic literature reviews (SLRs) are essential for health research and decision-making.
- Traditional SLRs are time-consuming and labor-intensive.
- Artificial intelligence (AI) and large language models (LLMs) offer potential for automating SLR processes.
Purpose Of The Study
- To conduct a systematic literature review on the cost-effectiveness of adult pneumococcal vaccination.
- To prospectively assess the performance of the AI-assisted review platform, ISLaR 2.0.
- To compare the AI platform's performance against expert human reviewers.
Main Methods
- A systematic literature review was performed on adult pneumococcal vaccination cost-effectiveness.
- The Intelligent Systematic Literature Review (ISLaR) 2.0 platform was utilized for review tasks.
- AI performance was evaluated against human expert reviewers in article screening and data extraction.
Main Results
- ISLaR 2.0 achieved high accuracy (0.87 for full-text screening, 0.86 for data extraction) and precision (0.88, 0.86).
- The platform demonstrated high sensitivity (0.91 screening, 0.98 data extraction) but lower specificity (0.79 screening, 0.42 data extraction), particularly with table data.
- ISLaR 2.0 reduced abstract and full-text screening time by over 90% compared to human reviewers.
Conclusions
- The ISLaR 2.0 platform shows significant potential to decrease reviewer workload in systematic literature reviews.
- Further development is needed to enhance the platform's overall performance and specificity, especially for complex data extraction.
- AI-assisted tools like ISLaR 2.0 are promising for improving the efficiency of health research processes.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.