Clinical Summarizer
Category: AI & ML
Node id: clinical-summarizer
Standards: HIPAA
Generates clinical summaries from patient records, FHIR bundles, or free text
What it does
- Comprehensive clinical summary generation
- Discharge summary with hospital course and follow-up
- SBAR handoff summary for clinical transitions
- Active problem list extraction
- Configurable summary length (short/medium/long)
- Multi-input support (free text, structured data, FHIR bundles)
Use cases
- Generate discharge summaries for hospitalized patients
- Create shift handoff reports in SBAR format
- Summarize patient records for referral letters
- Extract active problem lists from clinical notes
Runtime
- Performance tier:
variable
Input ports
| Name | Type | Required | Description |
|---|---|---|---|
clinical_text | text | No | Free-text clinical notes |
patient_data | json | No | Structured patient data |
fhir_bundle | json | No | FHIR Bundle resource |
Output ports
| Name | Type | Required | Description |
|---|---|---|---|
result | json | No | Main result containing summary text and summary_type |
success | boolean | No | Whether the operation succeeded |
Settings
Basic
| Setting | Name | Type | Required | Default | Allowed values | Description |
|---|---|---|---|---|---|---|
| Model | model | string | No | gpt-4o | Model name to use | |
| Summary Type | summary_type | enum | No | comprehensive | comprehensive, discharge, handoff, problem_list | Type of clinical summary |
| Max Length | max_length | enum | No | medium | short, medium, long | Target summary length |
Auth
| Setting | Name | Type | Required | Default | Allowed values | Description |
|---|---|---|---|---|---|---|
| LLM Endpoint | llm_endpoint | string | Yes | Base URL of the LLM API (e.g. https://api.openai.com/v1) | ||
| API Key | api_key | secret | No | Bearer token for the LLM endpoint |
Advanced
| Setting | Name | Type | Required | Default | Allowed values | Description |
|---|---|---|---|---|---|---|
| Max Tokens | max_tokens | number | No | 4096 | Maximum tokens for LLM response |
Errors
| Code | Description |
|---|---|
LLM_ENDPOINT_REQUIRED | No LLM endpoint configured |
NO_INPUT_DATA | No clinical text, patient data, or FHIR bundle provided |
LLM_REQUEST_FAILED | HTTP request to LLM endpoint failed or returned error |
Runtime notes
- At least one of clinical_text, patient_data, or fhir_bundle must be provided — all three can be combined for richer summaries
- result.summary is the generated text; result.summary_type echoes the configured type for downstream routing
- Handoff summary uses SBAR format (Situation, Background, Assessment, Recommendation) — ideal for shift change workflows
- Temperature is fixed at 0.3 for factual consistency; override via max_tokens config only
See Build any workflow for how to put nodes together and Ports, types and data flow for how port types connect.
¿Algo no está claro? Envíe sus comentarios sobre la documentación por correo electrónico..