Document Panel
The Document Panel displays the full clinical document with extracted concepts highlighted in context. It serves as the primary interface for reviewing and interpreting patient documentation.
Use the Document Panel when you need to:
- Read full clinical notes
- Identify key concepts within the document
- Validate extracted information
- Explore relationships between concepts and events
- Select a document from the Patient Document Timeline
- Review highlighted concepts within the document
- Hover over concepts to inspect confidence
- Click concepts to view detailed information
Each highlighted segment represents an extracted concept within the document.
Concepts are visually highlighted directly within the document text.
What’s happening: The system identifies and marks relevant clinical entities (e.g., medications, conditions).
Why it matters: Users can quickly locate important information without reading the entire document line-by-line.
Hovering over a highlighted concept displays its confidence level.
What’s happening: Each concept is assigned a confidence score based on extraction accuracy.
Why it matters: Helps users assess how reliable a concept identification is before using it for analysis.
Clicking a concept (in the document or Concept List) displays detailed information.
What’s happening: The selected concept becomes active and may reveal:
- Additional metadata
- Related mentions
- Contextual relationships
Why it matters: Enables deeper exploration beyond surface-level highlighting.
- Filters adjust which concepts are emphasized or de-emphasized
- Concepts below the confidence threshold may appear greyed out
- Group filtering limits visible concept categories
Why it matters: Allows users to tailor the document view to their analysis needs.
- Concepts are extracted using NLP-based recognition
- Multiple mentions of the same concept may appear throughout the document
- Highlighting is dynamic and responds to filter changes
- Selecting a concept may update other UI components
A user reviewing a clinical note can:
- Quickly identify mentions of medications or conditions
- Hover to verify confidence levels
- Click a concept to explore additional details
- Use filters to focus only on high-confidence concepts
- Some concepts may be incorrectly identified or missed
- Overlapping or dense highlights may occur in complex documents
- Confidence scores are estimates, not guarantees of accuracy
- Greyed-out concepts are still present but deprioritized
- See Filters for controlling concept visibility
- See Patient Document Timeline for document selection
- See Event Document Flow for how document concepts relate to events