Data Extraction
Automated Data Extraction for Systematic Reviews
Extract key information from studies with AI, automatically linked to highlighted sources in your PDFs for easy verification.
Effects of Metformin on Glycemic Control in Adults with Type 2 Diabetes: A Multi-Centre Study
Johnson A, Smith B, Patel C, Williams D
Journal of Clinical Endocrinology & Metabolism, Vol 109 (3), 2024
Abstract
Background: This randomized controlled trial1 evaluated the efficacy of metformin monotherapy in newly diagnosed patients with type 2 diabetes across 12 clinical sites in North America and Europe.
Methods: A total of 245 adults aged 18–65 with confirmed type 2 diabetes2 were enrolled and randomized 1:1 to metformin 1 000 mg daily or matched placebo for 12 weeks. The primary endpoint was change in HbA1c from baseline.
Results: The primary endpoint of HbA1c reduction at 12 weeks3 showed significant improvement in the treatment group (−1.2% vs −0.3%, p<0.001). Secondary endpoints including fasting glucose and body weight also favoured metformin.
Study Design
Randomized Controlled Trial
Sample Size
245 participants
Population
Adults 18–65, Type 2 Diabetes
Primary Outcome
HbA1c reduction at 12 wk
Manual Extraction Is Slow and Error-Prone
Data extraction is one of the most labor-intensive stages of a systematic review. Reviewers must read full-text articles, identify relevant data points across tables, figures, and text, and manually enter them, often across dozens or hundreds of studies.
How It Works
From Form to Finished Extract in Four Steps
Define Your Extraction Form
Create custom extraction forms that match your protocol. Define the exact fields you need: study characteristics, patient demographics, interventions, outcomes, and more.
AI Reads Your PDFs
EasySLR converts PDF documents into structured text and uses retrieval-augmented generation (RAG) to extract data from relevant passages. Each extracted value is linked to the highlighted source text in the original PDF.
Review and Verify
AI-extracted values are suggestions, not final answers. Reviewers verify each field, check the highlighted source, and approve or correct the extraction. Every decision is recorded in the audit trail.
Export Your Data
Export your extracted data in structured formats for analysis. All extraction decisions are exportable for external audit.
Key Capabilities
Purpose-built tools for reliable, auditable data extraction.
Custom Extraction Forms
Define any fields you need for your review.
Source-Linked Extraction
Every extracted value linked to highlighted text in the PDF.
Human Verification
Reviewers approve or correct every AI suggestion.
Audit Trail
Complete record of who extracted what, with timestamps.
Quality Control
QC mode for sampling and override of extractions.
Structured Export
Export data in formats ready for analysis.
Grounded in Your Documents, Not Invented
EasySLR uses retrieval-augmented generation (RAG) to ground extraction outputs in the actual text of your PDFs. This reduces the risk of unsupported outputs compared to pure generative AI. However, human verification remains essential. AI extraction is a starting point, not a final answer.
Trust & Compliance
Enterprise-Grade Security
Your research data deserves the highest level of protection. We maintain independently audited certifications and established data protection controls so you can focus on your review.
Explore More Features
AI-Assisted Screening
Screen titles, abstracts, and full texts with configurable AI modes.
Learn moreAll Features
See the complete platform: workflows, analytics, collaboration, and more.
Learn moreSystematic Review Software
Complete overview of EasySLR as AI-powered systematic review software.
Learn more
Claim your
free project now
No credit card required