Developing a fully automated evidence synthesis tool for identifying, assessing and collating the evidence

Area of Science:

• Medical Informatics
• Computational Biology
• Health Services Research

Background:

• Evidence synthesis is crucial for evidence-based medicine but is often labor-intensive and slow.
• Current methods struggle to incorporate all relevant trials, leading to outdated or incomplete syntheses.
• Automation technology offers a potential solution to these limitations.

Purpose of the Study:

• To describe a fully automated evidence synthesis system for intervention studies.
• To demonstrate a novel approach for identifying, assessing reliability, and collating evidence.
• To provide a publicly available tool for real-time evidence synthesis.

Main Methods:

• Utilized machine learning, natural language processing, and rule-based systems for automated evidence identification and assessment.
• Developed a system to collate evidence and estimate the relative effectiveness of interventions.
• Integrated modern visualization techniques for presenting synthesized results.

Main Results:

• Successfully developed and described the first publicly available, fully automated evidence synthesis system.
• The system can identify relevant evidence, assess its reliability, and synthesize findings on the fly.
• Results are presented using advanced visualization techniques.

Conclusions:

• Automated evidence synthesis systems can overcome the labor-intensive nature of traditional methods.
• This novel tool accelerates the process of evidence synthesis for intervention studies.
• The system offers a valuable resource for evidence-based medicine, providing up-to-date evidence mapping.