Leasey
Healthcare

Multimodal patient intake with schema-constrained extraction

Intake pipeline that parses forms, IDs, and clinical photos into a clean patient record with confidence scoring and human review queue.

>95%
field-level extraction accuracy
~90%
reduction in manual data entry
Days
to backfill years of archives

Challenge

Intake forms came in as PDFs, photos, and faxed scans. Manual entry was expensive and error-prone, and downstream systems needed strictly typed data.

Approach

  • Layout-aware parsing + vision-language extraction with schema validation
  • Cross-field business-rule validators for clinical coherence
  • Confidence scoring routes low-confidence items to a review queue
  • Backfill of historical archives in parallel batches

Outcomes

Clean
typed records into downstream systems
Review UI
on anything below confidence threshold

Stack

AWS Textract
Claude Vision
OpenAI Structured Outputs
AWS Lambda
Build something like this

Book a discovery call. We'll scope the right engagement for your version of this problem.

Book a call

Ready to accelerate your tech growth?

Schedule your free consultation today and let's discuss how we can help your business scale efficiently.

Tech growth illustration
Ready when you are

Let’s ship your AI system.

Whether you’re scoping a new LLM product, hardening an existing one, or standing up the infra behind it — we’ll map the shortest path to production.

Email the teamOther ways to reach us