Aller au contenu principal

RLM Clinical Notes

Category: AI & ML
Node id: rlm-clinical-notes
Maturity: beta
Risk tier: B

Standards: HIPAA

Single-call LLM presets for clinical notes: SOAP generation, structuring, entity extraction, summarization, coding suggestions. One chat completion per run — use rlm-query for the recursive RLM loop

What it does​

  • SOAP note generation from encounter transcripts or data
  • Clinical note structuring into standard formats
  • Entity extraction (diagnoses, medications, procedures, vitals, lab results)
  • Clinical note summarization
  • ICD-10-CM and CPT code suggestion with confidence levels
  • Configurable note format (SOAP, H&P, etc.)

Use cases​

  • Generating SOAP notes from physician-patient encounter transcripts
  • Structuring unstructured clinical notes into standard formats
  • Extracting structured entities from free-text clinical documentation
  • Suggesting billing codes from clinical documentation
  • Summarizing lengthy clinical notes for quick review

Runtime​

  • Performance tier: variable

Input ports​

NameTypeRequiredDescription
clinical_notetextNoExisting clinical note for processing
encounter_datajsonNoEncounter data for note generation
transcripttextNoEncounter transcript for note generation

Output ports​

NameTypeRequiredDescription
resultjsonNoClinical note result containing output, operation, note_format, and http_status
successbooleanNoWhether the operation succeeded

Settings​

Basic​

SettingNameTypeRequiredDefaultAllowed valuesDescription
LLM Endpointllm_endpointstringYesBase URL of the LLM API
ModelmodelstringNogpt-4oModel name to use
OperationoperationenumNogenerate_soapgenerate_soap, generate_note, structure, extract_entities, summarize, code_suggestClinical notes operation
Note Formatnote_formatstringNoSOAPClinical note format

Auth​

SettingNameTypeRequiredDefaultAllowed valuesDescription
API Keyapi_keysecretNoAPI key for the LLM provider

Advanced​

SettingNameTypeRequiredDefaultAllowed valuesDescription
TemperaturetemperaturenumberNo0.3min 0; max 2Sampling temperature
Max Tokensmax_tokensnumberNo4096Maximum tokens to generate

Errors​

CodeDescription
ENDPOINT_REQUIREDllm_endpoint is not configured
INPUT_REQUIREDRequired input missing for the selected operation
UNSUPPORTED_OPERATIONOperation must be generate_soap, generate_note, structure, extract_entities, summarize, or code_suggest
LLM_REQUEST_FAILEDLLM API call failed during clinical note processing
HTTP_ERRORLLM endpoint returned an HTTP error status

Runtime notes​

  • generate_soap and generate_note require either transcript or encounter_data input
  • structure, extract_entities, summarize, and code_suggest require clinical_note input
  • The output field is free-text LLM content; for extract_entities it may contain JSON embedded in text
  • LLM timeout is 120s; large notes may need higher max_tokens config

See Build any workflow for how to put nodes together and Ports, types and data flow for how port types connect.

Quelque chose n'est pas clair ? Envoyez vos commentaires sur la documentation par courriel.