Home/ Portfolio/ Fern Healthcare
Fern Healthcare · Healthcare · 2024 AI Mobile Healthcare

37 minutes saved per session. Notes in three minutes flat.

Speech-to-structured-note, with a custom model tuned on Pakistani clinical terminology and specialty-specific templates.

37min saved per session
94% note accuracy
8 specialties supported

Doctors at Fern were spending 35-40 minutes after each clinic session completing notes in their EMR. With 20 patients a day, that's an extra full workday of typing each week.

We built a real-time transcription and structuring system tuned on Pakistani Urdu-English medical terminology with specialty templates for GP, orthopaedics, and cardiology. The doctor speaks; the system produces a structured note. They review and sign off in three minutes.

The mobile app works fully offline during a session and syncs to the EMR via HL7 FHIR when connected. Notes are structured as SOAP (Subjective, Objective, Assessment, Plan) by default, with custom templates per specialty.

This project is in production but not publicly demoed. Details available under NDA — reach out.

Related work

Your project next

We keep the details confidential — you get the full story in a call.

Most of what we've built is under NDA. If you see a problem here that looks like yours, let's talk.

Chat on WhatsApp