Skip to main content

Getting Started with Castor Catalyst

Updated

Castor Catalyst is a clinical data review platform that simplifies how research teams handle source data. 

It allows you to ingest data from structured EMR records and PDF documents, review extracted data points via AI-powered pipelines, and submit them directly to Castor CDMS.

Note: This is an additional feature which incur costs. Please reach out to support@castoredc.com for inquiries regarding cost and activation

1. Account Activation and Sign-Up

Before accessing Catalyst, you must activate your account through Castor EDC (if you already have an account in Castor EDC/CDMS then you can skip this step)

  • Check your Email: Look for an invitation email from "Castor EDC" (no-reply@castoredc.com).
    Check your spam folder if you don't see it.

  • Register: Click the link in the email to go to the 'Welcome' page.

  • Set Credentials: Enter your name and create a password. Your password must include:

    • At least 8 characters.

    • 1 uppercase character.

    • 1 lowercase character.

    • 1 numeric character.

  • Finalize: Agree to the Terms of Use and click Sign Up.

Note: After signing up, you may see a login page for Castor EDC; however, you do not need to log in there to use Catalyst (see step 2)

2. Logging into Castor Catalyst

Users log in exclusively via their Castor Identity account.

  1. Navigate to your study's specific Catalyst URL (e.g., {base_url}.catalyst.castoredc.com).

  2. Click the Login via Castor Identity button.

  3. Enter the email and password you created during the activation step and click Sign In.

3. Selecting Context and Uploading Documents

Once logged in, you will only see the studies and sites you have permission to access.

Step 1: Select Participant and Timepoint

  • Study/Participant: Select the relevant study and participant ID from the dropdown menus.

  • Timepoint: Select the specific clinical timepoint (e.g., Baseline, Year 1) relevant to the data you are providing.

Step 2: Upload PDF Source

Catalyst uses a Document Processing Pipeline for automated AI-powered data extraction from PDFs.

  • Upload: Click Browse or drop a PDF file at the bottom of the page or drag and drop the file directly into the window.

  • Confirm: A dialog will appear to verify the Study, Participant, and Timepoint. Click Upload to begin the automated verification.

4. Data Review and Submission

All processed data—whether from EMR (FHIR) or PDF uploads—is presented in a Unified Review Interface.

  • Visual Evidence: For PDF sources, data points are visually highlighted on the original document using a bounding box.

  • Review Policy: A 100% review policy is enforced for document data; no data is submitted to the CDMS without explicit user approval.

  • Manual Overrides: You can approve AI-extracted values or manually override them. The system supports various field types including Numeric, Date, Dropdown, and Checkboxes.

  • Validation: Data is validated against CDMS fields before submission to prevent duplicates or overwriting.

5. Troubleshooting: Failed Uploads

If the participant ID or timepoint extracted from the document does not match your selection, the upload will fail.

  • The timepoint status will change to Invalid.

  • Specific error details will be listed at the bottom of the page.

  • Click Clear and try a new file to restart the process.

Support

If you encounter technical issues, please contact the support team: