Toward systematic review automation: a practical guide to using machine learning tools in research synthesis.

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Evidence Synthesis Methodologies

Background:

• Systematic reviews are crucial for evidence-based practice but are time-consuming.
• Emerging technologies aim to automate key steps like searching, screening, and data extraction.
• Lack of practical guidance hinders the adoption of these automation tools.

Discussion:

• This guide overviews machine learning (ML) methods for expediting evidence synthesis.
• It addresses the practical implementation, strengths, and weaknesses of ML tools.
• Guidance is provided for systematic review teams on integrating these technologies.

Key Insights:

• Machine learning offers significant potential to speed up systematic review production.
• Understanding the readiness and limitations of specific ML tools is essential for effective use.
• Practical implementation strategies are crucial for successful automation in evidence synthesis.

Outlook:

• Further research and development in ML for systematic reviews are expected.
• Standardized guidelines for using automated tools will likely emerge.
• Increased adoption of ML promises more efficient and timely evidence synthesis.