Negated symptom
Model input
Document excerpt
Clinical note
Synthetic note: Denies chest pain. Reports intermittent dizziness.
Clinical document extraction
Extract documented clinical facts, structure report findings, and identify patient and provider roles.
Choose an example
Abstract explicitly documented clinical facts from synthetic notes.
Step 01
Define and Align
Choose an example, review the proposed output, and annotate the fields that need to change. Your guidance defines the learned rules.
Output schema
- Chest painYes / No / Unknown
Whether current chest pain is explicitly documented.
- true
- false
- null
- Asthma statusChoose one
Preserve whether asthma is current, historical, or denied.
- current
- historical
- negated
- not_documented
- Pneumonia statusChoose one
Preserve the certainty of the documented finding.
- confirmed
- possible
- negated
- not_documented
Collecting feedback · 1/3
Abstract explicitly documented clinical facts from synthetic notes.
Validation accuracy
—
Example 1 of 3
3 annotations saved
Sutro can now generalize the rules based on your feedback.
Step 02
Model Sweep
Compare quality and cost on your examples, then choose a model or use Sutro’s default.
Compare models on expert agreement and cost. The interactive chart uses illustrative measurements.
Illustrative sweep · select Value or Best to use that model in your function.
Step 03
Deployment
Call Sutro Functions like any other software. Swap the model as soon as something more performant is released.
Selected modelWaiting for sweep results.
Real-time
import sutro as so
input_data = {
"text": "Synthetic note: Denies chest pain. Reports intermittent dizziness."
}
result = so.run_function(
"clinical-data-abstraction",
input_data,
)curl --fail-with-body -X POST "${SUTRO_API_URL%/}/v1/functions/clinical-data-abstraction/run" \
-H "Authorization: Key $SUTRO_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"input": {
"text": "Synthetic note: Denies chest pain. Reports intermittent dizziness."
}
}'Batch
import sutro as so
job_id = so.batch_run_function(
name="clinical-data-abstraction",
data="clinical_data_abstraction.csv",
)
results = so.await_job_completion(job_id)curl --fail-with-body -X POST "${SUTRO_API_URL%/}/v1/batch-inference" \
-H "Authorization: Key $SUTRO_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "clinical-data-abstraction",
"inputs": [
{
"text": "Synthetic note: Denies chest pain. Reports intermittent dizziness."
}
]
}'