Treatment Planning
Category: AI & ML
Node id: treatment-planning
Standards: HIPAA
AI-powered evidence-based treatment plan generation
What it does
- Evidence-based treatment plan generation
- Multi-provider LLM support (OpenAI, Anthropic)
- Alternative treatment options with trade-offs
- Cost estimation for treatment options
- Short-term and long-term planning horizons
- Patient preference and insurance awareness
Use cases
- Generate treatment plans from confirmed diagnoses
- Compare alternative treatment approaches
- Estimate treatment costs for patient counseling
- Create guideline-based care recommendations
Runtime
- Performance tier:
variable
Input ports
| Name | Type | Required | Description |
|---|---|---|---|
diagnosis | text | Yes | Diagnosis information |
patient_context | json | No | Patient context (sub-keys: patient_data, medications, allergies, lab_results, conditions, preferences, insurance, age, gender) |
Output ports
| Name | Type | Required | Description |
|---|---|---|---|
result | json | No | Main result containing treatment_plan, summary, timeline, alternatives, cost_estimates, and disclaimer |
metadata | json | No | Execution metadata including tokens, treatment_count, essential/recommended/optional counts |
success | boolean | No | Whether the operation succeeded |
Settings
Basic
| Setting | Name | Type | Required | Default | Allowed values | Description |
|---|---|---|---|---|---|---|
| Provider | provider | enum | No | openai | openai, anthropic | LLM provider |
| Model | model | string | No | "" | Model name (auto-selected if empty) | |
| Guideline Level | guideline_level | enum | No | standard | standard, evidence-based, experimental | Clinical guideline level |
| Timeframe | timeframe | enum | No | short-term | short-term, long-term, both | Treatment plan timeframe |
| Include Alternatives | include_alternatives | boolean | No | true | Include alternative treatment options | |
| Include Cost Estimate | include_cost_estimate | boolean | No | false | Include rough cost estimates |
Auth
| Setting | Name | Type | Required | Default | Allowed values | Description |
|---|---|---|---|---|---|---|
| API Key | api_key | secret | No | API key for the LLM provider | ||
| Base URL | base_url | string | No | "" | Custom API base URL |
Advanced
| Setting | Name | Type | Required | Default | Allowed values | Description |
|---|---|---|---|---|---|---|
| Temperature | temperature | number | No | 0.3 | min 0; max 1 | Sampling temperature |
| Max Tokens | max_tokens | number | No | 4000 | min 1 | Maximum tokens for response |
Errors
| Code | Description |
|---|---|
DIAGNOSIS_REQUIRED | No diagnosis text provided in inputs |
INVALID_PROVIDER | Provider must be openai or anthropic |
LLM_REQUEST_FAILED | LLM API call failed or returned error |
PARSE_ERROR | Failed to parse LLM response as structured treatment plan |
Runtime notes
- result always includes a disclaimer string — do not suppress it when displaying to users
- patient_context sub-keys map internally: patient_data, medications, allergies, lab_results, conditions, preferences, insurance
- If LLM response cannot be parsed as JSON, a fallback single-treatment plan is returned with the raw text as rationale
- Latency is variable and depends on LLM provider; plan for 5-30s execution time
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.