Documents

The Hub lets you upload patient documents and use AI to search and ask questions about their contents.

Uploading Documents

Drag and drop files onto the upload area, or click to browse your computer. Supported formats include PDFs and images (max 50MB). A progress bar shows upload status, and documents appear in your list once processing completes.

The document count in the header updates as you add files. Documents are associated with the current patient context.

Processing Pipeline

After uploading, documents go through an automated processing pipeline. You’ll see real-time status updates as each stage completes:

Step	Stage	What Happens
1	Preparing	Document is queued for processing
2	Reading	OCR extracts text from every page
3	Extracting	AI identifies medications, conditions, labs, vitals, demographics
4	Matching patient	Extracted demographics are matched to existing patients
5	Creating patient	A new patient record is created if no match is found
6	Summarizing	AI generates a clinical summary of the document
7	Finalizing	Extracted data is saved and indexed for search

Note

Most documents complete processing in under 30 seconds. Multi-page documents may take longer during the reading stage.

Extracting Patient Data

When uploading documents, you can automatically extract clinical data and add it to the patient’s record. Select which data types to extract:

Category	What’s Extracted
Medications	Drug names, doses, frequencies
Conditions	Diagnoses and ICD codes
Allergies	Allergens and reactions
Labs	Test values and dates
Vitals	BP, heart rate, temperature, weight
Demographics	Name, DOB, MRN

After processing, extracted data is automatically merged into the patient’s clinical record. The system deduplicates entries to avoid adding information that already exists.

Experimental

Patient data extraction is experimental. Review extracted data for accuracy before relying on it for clinical decisions.

Viewing Documents

Click any document to open the viewer. Multi-page documents display one page at a time with navigation controls to move between pages or jump to a specific page number.

Documents render at high resolution and scale responsively to fit your screen. When you search or ask questions, relevant sections are highlighted directly on the document pages so you can see exactly where information was found.

AI-Powered Search

Ask questions about document contents using the AI chat. Type a question like “What was the patient’s blood pressure at the last visit?” or “Summarize the discharge instructions.” The AI searches through the document and provides an answer with source references.

Search across documents to find specific information. Results show relevant passages along with which document and page they came from. Click any result to jump directly to that location in the document.

Scope your search to either the current document or all documents for this patient. Toggle between scopes depending on whether you need information from a specific document or want to search everything.

Tip

For best results, ask specific questions. “What medications were prescribed?” works better than “Tell me about this document.”

Next Steps

Patient Details

View clinical data →

Managing Patients

Organize your patient list →