Synthetic Pediatric Patient Generator

A comprehensive system for generating synthetic pediatric patients with complete medical histories and educational cases for medical learners.

Features

Patient Generation

• Single Patient Creation: Generate individual patients with complete longitudinal histories
• Bulk Generation: Create patient panels for educational cohorts
• Realistic Timelines: Birth to present with age-appropriate milestones and events

Medical History Components

• Vitals: Age-appropriate vital signs with growth over time
• Laboratory Results: Common and specialty labs with normal/abnormal values
• Imaging Studies: X-rays, ultrasounds, CT, MRI with findings
• Growth Charts: WHO/CDC growth percentiles tracked over time
• Immunizations: CDC schedule compliance with catch-up scenarios
• Developmental Milestones: Motor, language, social, cognitive
• Medical Events: Illnesses, procedures, hospitalizations, medications

Case Generation

• Epidemiologically Accurate: Seasonal patterns, geographic prevalence
• Curriculum-Mapped: Aligned with pediatric core topics and diagnoses
• Difficulty Levels: Beginner, intermediate, advanced
• Complete Presentations: Chief complaint, HPI, physical exam, working differential

Panel Management

• Learner Assignment: Individual or group patient panels
• Case Serving: On-demand case delivery from patient histories
• Progress Tracking: Curriculum coverage and diagnostic reasoning

Architecture

peds/
├── src/
│   ├── models/           # Data models for patients, cases, curriculum
│   ├── generators/       # Patient and case generation logic
│   ├── engines/          # Epidemiological and clinical reasoning engines
│   ├── panels/           # Panel management and case serving
│   └── api/              # FastAPI endpoints
├── data/
│   ├── curriculum/       # Pediatric core topics and diagnoses
│   ├── epidemiology/     # Seasonal and geographic disease patterns
│   ├── growth_charts/    # WHO/CDC reference data
│   └── immunizations/    # CDC vaccination schedules
├── tests/
└── examples/

Getting Started

Quick Install

cd peds
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Generate Your First Patient

# Generate a 5-year-old patient
python -m src.cli patient --age 5

# Save to file
python -m src.cli patient --age 5 --output my_patient.json

# Generate a panel of 20 patients
python -m src.cli panel --count 20 --output patients/

# Create an educational case
python -m src.cli case --age 3 --difficulty beginner --output case.json

# See all available commands
python -m src.cli --help

Documentation

• QUICKSTART.md - Get started in 5 minutes
• USAGE.md - Comprehensive usage guide with examples
• ARCHITECTURE.md - System design and technical details

Use Cases

1. Medical Education: Create realistic patient panels for clinical reasoning practice
2. Curriculum Design: Ensure comprehensive topic coverage
3. Assessment: Generate cases for exams and competency evaluation
4. Research: Test diagnostic algorithms and clinical decision support tools

Technology Stack

• Python 3.11+
• FastAPI: REST API for patient and case generation
• Pydantic: Data validation and schema management
• SQLAlchemy: Database ORM for persistence
• LangGraph: Orchestrate complex generation workflows
• Anthropic Claude: LLM-powered clinical reasoning and narrative generation