Mini Data Extraction in EasySLR

Mini Data Extraction in EasySLR

The Mini Data Extraction feature in EasySLR allows teams to extract a focused set of predefined fields during the screening process. It's ideal when only high-level or limited data is required before full-scale data extraction begins.

How to Set Up Mini Data Extraction
  1. Click on Settings.
  2. Open the Mini Data Extraction tab.


  1. Set Up Fields
    • Click Add Field to define your data points.
    • Enter a Field Name.
    • Choose an Input Type:
      1. Text
      2. Multi-select
      3. Yes/No
    • Mark fields as Required, if necessary.

  • Repeat to add as many fields as needed for:
    1. Title–Abstract screening
    2. Full-Text screening
  • Choose whether a conflict applies based on the data filled (useful during the QC).
  • Click Update to save your form.
 
Enabling AI for Mini Data Extraction
  • Navigate to: AI Copilot → Title Abstract/ Full Text Mini DE AI
  • Enable AI extraction


Rerun AI by Clicking on Rerun AI at Title Abstract/Full Text Mini Data Extraction AI Suggestions Refresh.

Choose the relevant option and Rerun AI.

You can further review, add, or update the AI-extracted data as needed.

AI-extracted data will be available in separate columns in the Excel file, along with the AI decisions.

How to Manually Add Data During Screening
  1. Go to Screening Stage
    • Navigate to Articles to Review in either the Title–Abstract or Full-Text screening tab.
  1. Enter Data
    • For each article, the Mini Data Extraction fields will appear.
    • Fill out the relevant information before submitting a decision.


  1. Exporting Your Data
    • Go to the Articles tab.
    • Click Download > Download All to Excel.
  • The exported file will include:
    • All Mini Data Extraction fields
    • Decisions by each reviewer

Functional Workflow: 

When Decisions Match:
  • If Reviewer 1 (R1) and Reviewer 2 (R2) decisions match:
    • The Final Decision and Mini Data Extraction values are finalised based on the hierarchy order of the reviewers.
When Decisions Conflict:
  • If R1 and R2 decisions differ:
    • The Final Reviewer will be presented with
      • Inputs from R1 and R2 for direct comparison.
    • Final Reviewer Actions:
        • Copy values directly from R1 or R2.
        • Edit copied values or input new values.
        • Save the data.
        • The data will be mapped against the final reviewer
QC:
  • The project owner can apply filters to refine the list of articles as needed.
  • Articles can be filtered based on the information entered in the Mini Data Extraction (Mini DE) fields.
  • Additionally, articles with mismatched reviewer inputs can be identified using the Mini DE Conflict filter (this can be set up from settings).
  • During the decision override, the project owner can view Mini DE data from all reviewers and make necessary edits or updates.

Tips
  • You can customise fields per screening stage.
  • Use Mini Data Extraction to quickly gather structured information early in your workflow.
Notes
  • Only project owners or users with edit access can modify Mini Data Extraction settings.
  • Data captured here complements, but does not replace, full data extraction.
  • AI outputs should be reviewed and validated

Need Help?
Reach out to us via the Live Chat option in EasySLR or email us at support@easyslr.com.


    • Related Articles

    • Direct PDF Upload to Data Extraction

      EasySLR allows you to start Data Extraction directly by uploading PDFs, without first completing Title–Abstract or Full-Text screening. This feature is designed for teams that already have finalised studies or want to begin extraction immediately, ...

    • How to set up and Extract Data in EasySLR?

      Overview Data extraction in EasySLR is designed to be flexible and easy to set up. You can configure your data extraction framework either: During project setup, or After completing Title–Abstract (TiAb) and Full-Text screening EasySLR provides ...

    • List of AI Models Used Across EasySLR

      EasySLR uses purpose-specific GPT models across different review stages to optimise accuracy, performance, and cost efficiency. Each model has been selected based on the complexity and nature of the task it supports, ensuring reliable AI assistance ...

    • Guide to tracking AI and data processing tasks in EasySLR

      Tasks triggered by user actions, such as AI processing or Data Extraction runs, continue to operate in the background. Depending on the size of data, these tasks may take some time to complete. Tracking the progress of these tasks gives you full ...

    • Report Generation in EasySLR

      The Report Generation feature in EasySLR helps you automatically generate structured, publication-ready reports based on your project data. It consolidates extracted data, screening outcomes, PRISMA flow, and key study summaries into a single, ...