Skip to content

Healthcare Graph Fundamentals

View Healthcare Graph Fundamentals Fullscreen

This visualization demonstrates the fundamental components of a graph—nodes and edges—using a simple healthcare scenario. It features a patient (Sarah Chen) interacting with various healthcare entities including a provider (Dr. Martinez), a diagnosis (Type 2 Diabetes), medication (Metformin), and a facility (City Hospital).

Overview

This interactive visualization illustrates what nodes (entities) and edges (relationships) are in graph theory by modeling a basic healthcare scenario. Each node represents a distinct entity in the healthcare system with its own unique properties, while edges show the relationships between these entities. This foundational example demonstrates how graph databases naturally model real-world healthcare interactions.

Features

Node Types

  • Patient (Blue, Circle): Represents individual patients in the healthcare system

  • Properties: name, age, mrn, admission_date

  • Provider (Green, Circle): Healthcare providers such as physicians and specialists

  • Properties: name, specialty, npi, department

  • Diagnosis (Orange, Circle): Medical diagnoses and conditions

  • Properties: name, icd10_code, category, severity

  • Medication (Purple, Circle): Medications prescribed to patients

  • Properties: name, generic_name, dosage, frequency

  • Facility (Yellow, Circle): Healthcare facilities and hospitals

  • Properties: name, type, address, bed_capacity

Interactive Features

  • Hover Tooltips: Hover over any node to see its type and properties in a tooltip
  • Click for Details: Click on a node to display full details in the right sidebar
  • Search: Use the search box to find nodes by name or property values
  • Filter by Type: Toggle checkboxes to show or hide specific node types
  • Select All/Unselect All: Bulk controls to quickly filter all node types
  • Force-Directed Layout: Automatic node positioning using physics simulation

Educational Insights

  • Graph Basics: A graph consists of nodes (entities) and edges (relationships). Nodes represent things (patient, provider, diagnosis), while edges represent how these things are connected (visited, diagnosed with, prescribed).

  • Heterogeneous Properties: Notice how each node type has completely different properties—patients have MRN and age, providers have NPI and specialty, diagnoses have ICD-10 codes. In a graph database, this flexibility is natural, while relational databases would require multiple tables or sparse schemas.

  • Directed Relationships: The arrows on edges show direction. For example, "VISITED_PROVIDER" flows from Patient to Provider, indicating the patient visited the provider (not vice versa). This directionality captures the semantic meaning of relationships.

Data Structure

The visualization uses a JSON data structure with nodes and optional edges:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
{
  "nodes": [
    {
      "id": 1,
      "label": "Example Node",
      "type": "NodeType",
      "properties": {
        "property1": "value1",
        "property2": "value2"
      },
      "shape": "circle",
      "color": "#ec4899",
      "size": 40
    }
  ],
  "edges": [
    {
      "from": 1,
      "to": 2,
      "label": "RELATIONSHIP_TYPE",
      "arrows": "to"
    }
  ]
}

Implementation Notes

  • Library: vis-network JavaScript library
  • Layout: Force-directed using Barnes-Hut algorithm
  • Icons: None (using colored shapes only)
  • Data File: healthcare-graph-fundamentals-data.json

Next Steps

After understanding this basic graph structure, you can:

  1. Add More Nodes: Expand the graph with additional patients, providers, or medications to see how the network grows
  2. Add More Relationships: Create new edge types like "WORKS_AT" (Provider to Facility) or "ALLERGIC_TO" (Patient to Medication)
  3. Explore Properties: Click on each node to see how different entity types store different kinds of information
  4. Filter and Search: Practice using the interactive controls to explore subsets of the graph
  5. Compare to Tables: Consider how you would model this same information in a relational database—how many tables would you need?