SOAP Notes

​

Overview

​

What is a SOAP Note?

​

Data Model

class SOAPNote(BaseModel):
    subjective: str
    objective: str
    assessment: str
    plan: str
    confidence_score: float = Field(ge=0.0, le=1.0)

​

Generation Process

​

Implementation

async def _generate_soap_note(self, state: MedicalAnalysisState) -> MedicalAnalysisState:
    """Generate SOAP note from case summary"""
    
    # Generate SOAP note using specialized prompt
    soap_response = await self.llm_manager.generate_response(
        system_prompt=SOAP_NOTE_PROMPT,
        user_input="Case Summary" + state["case_summary"].model_dump_json()
    )
    
    # Create structured SOAP note
    soap_note = SOAPNote(
        subjective=soap_response.get("subjective", ""),
        objective=soap_response.get("objective", ""),
        assessment=soap_response.get("assessment", ""),
        plan=soap_response.get("plan", ""),
        confidence_score=soap_response.get("confidence_score", 0.8)
    )
    
    state["soap_note"] = soap_note
    state["processing_stage"] = "soap_note_generated"
    return state

​

SOAP Note Components

​

Subjective Section

• Chief Complaint: Primary reason for visit
• History of Present Illness: Timeline and progression of symptoms
• Patient-Reported Symptoms: Description in patient’s own words
• Relevant Medical History: Past conditions, surgeries, medications
• Social History: Lifestyle factors affecting health

​

Objective Section

• Vital Signs: If available in documentation
• Physical Examination Findings: From clinical notes
• Laboratory Results: Processed from uploaded lab files
  • Blood counts, chemistry panels
  • Urinalysis results
  • Microbiology cultures
• Radiology Findings: From vision AI analysis
  • X-ray interpretations
  • CT/MRI findings
  • Ultrasound results

​

Assessment Section

• Clinical Interpretation: What the data means clinically
• Working Diagnosis: Most likely diagnosis based on evidence
• Differential Considerations: Other possible diagnoses
• Severity Assessment: Acute vs chronic, mild vs severe
• Prognosis: Expected outcome and course

​

Plan Section

• Diagnostic Tests: Additional investigations needed
• Treatment Recommendations: Medications, procedures, therapies
• Patient Education: Instructions and precautions
• Follow-up: When to return, monitoring requirements
• Referrals: Specialist consultations if needed

​

Example SOAP Note

{
  "subjective": "31-year-old female patient Jane Daniel presents with respiratory symptoms. Patient reports progressive shortness of breath and persistent cough over the past week. No fever reported initially. Patient works in healthcare setting with potential exposure to infectious agents.",
  
  "objective": "Laboratory findings show elevated WBC count at 12,500/μL (normal 4,000-11,000), indicating possible infection. CRP elevated at 45 mg/L (normal <10). Chest X-ray reveals bilateral interstitial infiltrates consistent with pneumonia. No pleural effusion noted. Oxygen saturation 94% on room air.",
  
  "assessment": "Acute community-acquired pneumonia, likely atypical pattern based on radiographic findings. Moderate severity based on vital signs and lab values. Patient is ambulatory and hemodynamically stable. Differential includes viral pneumonia, mycoplasma pneumonia, or early bacterial pneumonia.",
  
  "plan": "1. Initiate empiric antibiotic therapy with azithromycin 500mg daily for 5 days. 2. Supportive care with adequate hydration and rest. 3. Monitor oxygen saturation at home; return if drops below 92%. 4. Follow-up chest X-ray in 6 weeks to confirm resolution. 5. Return to clinic in 3 days or sooner if symptoms worsen. 6. Work restrictions until afebrile for 24 hours.",
  
  "confidence_score": 0.85
}

​

Input Data Flow

​

Data Integration

case_context = {
    "patient_info": "Name: Jane Daniel, Age: 31, Gender: Female",
    "doctor_notes": "Patient presents with respiratory symptoms",
    "lab_summaries": "Elevated WBC 12,500; CRP 45 mg/L",
    "radiology_summaries": "Bilateral interstitial infiltrates on CXR"
}

​

Confidence Scoring

​

Confidence Interpretation

• 0.9 - 1.0: Excellent documentation, all sections well-supported
• 0.8 - 0.9: Good documentation, minor gaps in data
• 0.7 - 0.8: Adequate documentation, some assumptions made
• Below 0.7: Limited data, requires manual review and enhancement

​

Accessing SOAP Notes

GET /cases/cases/{case_id}

Response:
{
  "case": {...},
  "files": [...],
  "ai_insights": {
    "soap_note": {
      "subjective": "...",
      "objective": "...",
      "assessment": "...",
      "plan": "...",
      "confidence_score": 0.85
    },
    "case_summary": {...},
    "primary_diagnosis": {...}
  }
}

​

Clinical Workflow Integration

​

Best Practices

​

Customization Potential

1. Prompt Engineering: Modify SOAP_NOTE_PROMPT for specialty-specific formats
2. Template Variations: Different templates for different specialties
3. Data Emphasis: Weight certain findings more heavily
4. Length Control: Adjust verbosity of each section

​

Use Cases

​

Limitations

​

Next Steps