Extracting data from stool test reports
Specialty labs report stool tests as multi-section panels, and each lab chooses its own marker list and graphics. Gastroenterology and integrative medicine practices receive one for every patient tested, often with a repeat test after treatment, as lab portal downloads or faxed pages.
Practices need the calprotectin, elastase and secretory IgA results, the short-chain fatty acid profile, and the specimen's collection date. An electronic health record charts each marker across repeat tests, so a clinician can see inflammation fall after treatment begins.
What makes stool test reports hard
A graphic in every row
Specialty stool panels draw a colored range bar in each result row, with a marker showing where the result falls against its interval.
Words and numbers in one column
Result columns mix numeric values with qualitative entries such as Negative, None or Soft, all under one Result heading.
One table, many sections
Digestion, inflammation, immunology and fatty acid markers sit in separate sections, each repeating its own Result, Unit and Reference Interval header row.
Parse, extract, verify
Parse
Parse turns the document into reading-order Markdown and a hierarchy of blocks (text, tables, figures), and grounds each block to its exact location on the page.
Extract
You define the fields in a JSON schema, and Extract returns their values in a consistent, structured format, each linked back to where it appears in the Markdown.
Verify
Each value links back to where it appears, down to the character range in the Markdown, and the page and bounding box in the source document, so every result can be verified.
One stool test report, end to end
About this document: A Stool Chemistries report that Mosaic Diagnostics publishes as a sample of its stool chemistry panel, analyzed by Doctor's Data. Three portrait pages: order, patient and collection boxes over 18 markers in six sections, then a guide to each marker and the lab's commentary. Page 1 is featured. Every patient and order detail is the lab's specimen data.
Processing objective: The objective is to identify the specimen behind the results: which panel, which order and when the sample moved through the lab. The specimen dates come back as printed, each grounded to its line in the header box. A clinician lines up repeat tests by collection date.
Demonstration fields: Report, Order number, Collected, Received, Reported
The document
3 of 3 pages
Page 1shown below
Page 2
Page 3
Where each value came from
report_title
{
"value": "Stool Chemistries",
"page": 1,
"range": {
"end": 17,
"start": 0
},
"box": {
"xmin": 0.07037,
"ymin": 0.0591,
"xmax": 0.24868,
"ymax": 0.07521
}
}order.order_number
{
"value": "999999-9999",
"page": 1,
"range": {
"end": 107,
"start": 88
},
"box": {
"xmin": 0.0596,
"ymin": 0.11536,
"xmax": 0.20522,
"ymax": 0.12859
}
}specimen_dates.collected
{
"value": "06/28/2026",
"page": 1,
"range": {
"end": 360,
"start": 334
},
"box": {
"xmin": 0.68066,
"ymin": 0.1313,
"xmax": 0.92871,
"ymax": 0.14522
}
}specimen_dates.received
{
"value": "07/01/2026",
"page": 1,
"range": {
"end": 386,
"start": 361
},
"box": {
"xmin": 0.67996,
"ymin": 0.14641,
"xmax": 0.92928,
"ymax": 0.1601
}
}specimen_dates.reported
{
"value": "07/07/2026",
"page": 1,
"range": {
"end": 412,
"start": 387
},
"box": {
"xmin": 0.68022,
"ymin": 0.1618,
"xmax": 0.92929,
"ymax": 0.17627
}
}How to try it yourself
Everything on this page is reproducible. The sample, its schema, and the Parse and Extract API outputs are published. You can repeat the same two calls from whichever surface you already build on.
- Claude Skilldescribe the job; the agent writes it
- CLIno code at all
- MCPexpose ADE docs as an agent tool
- REST APIany language
- Pythonlandingai-ade
- TypeScriptlandingai-ade
Documents into data
Accurate
The best accuracy on the difficult documents other tools get wrong, such as messy scans, complicated tables, checkboxes and handwriting.
Auditable
Every answer cites its sources. Values come back with an audit trail (page, bounding box, character range) so a human can review them.
Agentic
Applies multi-step reasoning, and plans before it acts. Handles documents it has never seen before and adapts to variations.
API-first
Composable APIs that drop into any tech stack and can be called from anywhere. Easy to build on and embed in your pipelines and products.
Affordable
Cost-competitive rates with complete price transparency. Spend controls before you run, and every job reports its exact cost.
Enterprise-grade
Built for production and scale from day one. Fully observable and monitored, running securely in our cloud or yours.
Built for high-volume, high-throughput production
- Zero data retention
- Priority rate limits
- Enhanced support
- VPC and on-prem deployment
Available on Team and Enterprise plans. VPC and on-prem deployment require an Enterprise plan.
