Case study 05 · Healthcare
Esperwise
An AI assistant that learns each doctor's own style. It interviews the patient before the visit, listens during it, and has the note ready to sign in under a minute.
Try it: follow one visit from start to finish, switch the transcript to Dutch, sign the note. The patient is fictional.
from the end of the visit to a note ready for the doctor to review.
less time on paperwork, the time doctors used to spend typing.
per doctor, trained on their own questions and their own way of writing.
notes saved to the record without the doctor's signature.
Doctors were typingwhile patients talked.
A big part of every appointment went on intake questions and notes instead of the patient. Generic AI tools did not help much: their notes did not sound like the doctor, so every one was rewritten by hand.
- The same intake questions asked at every visit
- Eyes on the keyboard during the consultation
- Notes finished at the end of the day, from memory
- AI drafts that had to be rewritten anyway
- Intake done by the patient before they sit down
- The consultation transcribed while the doctor listens
- A note ready to review in under a minute
- Written the way that doctor writes
The assistant writes.The doctor signs.
Tap any step to see what happens there. The doctor stands between the AI and the patient record, so nothing is saved that a doctor has not read.
Interview. The patient chats with the assistant on their phone before the appointment. It asks the doctor's questions and follows up when an answer needs more detail.
Same visit.Three different doctors.
Doctors do not write alike. One wants three lines of shorthand with the plan first. Another writes in full sentences because the note becomes a letter to the GP. Another writes in Dutch, in SOAP format.
A note in the wrong style gets rewritten, and then the time saved is lost. So each doctor's assistant learns from their own example notes, and drafts the way they would.
PLAN: XR R knee. R/v 2/52.
PC: R knee pain 3/52, twisting inj (football).
Hx: swelling post-walking, 1x locking.
O/E: medial joint line tender.
One assistant per specialty.One place for the practice.
Around the assistant sits everything a clinic runs on: patients, appointments, scanned documents and transcripts, with each person seeing only what their role allows.
A specialist for each clinic
Each assistant asks the questions that specialty needs, and politely steers patients back when they drift off topic.
Patients, appointments, documents
- Patient records with intake and visit history
- Appointment calendar by day, week or month
- Scanned letters and results, searchable
- Satisfaction and usage per assistant
Each role sees its own part
| Doctor | Nurse | Desk | |
|---|---|---|---|
| Create assistants | ✓ | – | – |
| Transcribe visits | ✓ | ✓ | – |
| Add patients | ✓ | ✓ | ✓ |
| Scan documents | ✓ | ✓ | ✓ |
| Book appointments | ✓ | ✓ | ✓ |
Example setup. Each practice decides its own.
Suggestions,never decisions.
An assistant that suggests next steps in a medical note is useful, and it is also the part that can do harm if it is presented wrongly. So suggestions are always marked as suggestions, kept apart from the note, and nothing reaches the record until a doctor signs.
The same rule shaped the intake: the assistant collects and organises what the patient says. It does not tell the patient what is wrong with them.
Map the visit
How each doctor runs a visit, what they ask, and how they write their notes.
Patient interview
Specialty assistants that follow each doctor's questions and stay on topic.
Listen and write
Live transcription in English and Dutch, turned into notes in each doctor's style.
The practice around it
Patients, appointments, documents, roles and permissions, and the review-and-sign step.
Private by design.
AI drafts stay drafts until a doctor reviews and signs them.
Doctors, nurses and front desk each see only what their role allows.
Designed with HIPAA in mind: patient data protected in storage and in transit.
Every patient, visit and note on this page is made up for the demo.
For the technical reader.
Speech-to-text for the consultation, in English and Dutch.
Each assistant retrieves from that doctor's own protocols and example notes.
The patient interview, note drafting, summaries and suggestions.
Structured notes that can be sent on to an electronic health record.
Common questions
What is Esperwise?
An AI assistant for specialist clinics that interviews patients before the visit, transcribes the consultation and drafts the note in each doctor's own style.
Which languages does it support?
English and Dutch, for transcription and notes.
How does it match each doctor's style?
Each doctor's assistant retrieves from their own protocols and example notes, so drafts read the way that doctor writes.
Is anything saved without the doctor?
No. AI drafts stay drafts until a doctor reviews and signs them.
More case studies
All case studies →VeraVoice
An AI receptionist that never lets a call ring out.
View case study →PreOp Screen
Getting patients ready for surgery, without the phone calls.
View case study →Retail Insights
From last week's spreadsheets to dashboards that are fresh every morning.
View case study →Syphon AI
Reads 200-page tenders, flags the deal-breakers, and cites every answer.
View case study →Recruiting NOW
An AI that interviews every applicant in German, at any hour, and explains every score.
View case study →Contract Blitz
Explains every clause in plain English, flags what bites, and tells you what to ask for.
View case study →Brands10x
An AI CEO, CFO, CMO, COO and Chief of Staff that plan together around the same goals.
View case study →WorkApply
Finds the jobs that fit, tailors every application, and applies for you.
View case study →BuildStudio
The design team trailer shops hire on Upwork, as one AI studio.
View case study →Your project, next.
Tell us what you want to build. You get a fixed, line-item quote after one call.
Start a project →