Healthcare · Case study
AI patient-intake and triage assistant for a telehealth clinic network
An AI intake and triage assistant for a telehealth clinic network. Patients describe their symptoms, an LLM pipeline built with LangChain asks adaptive follow-up questions, classifies urgency and the likely department, and gives doctors a ranked summary on a dashboard with history, flags and exportable reports.
The challenge
Patients describe symptoms in their own words, and every intake had to be read by a clinician before anyone could judge how urgent it was or which department should take it. The network needed that first pass done the same way every time, before a doctor opened the case.
Our approach
We built an LLM intake pipeline with LangChain on OpenAI models. It interprets the patient's description, adapts the follow-up questions to the answers, classifies urgency and the probable department, and hands the doctor a ranked summary instead of a raw form.
What we built
- Symptom intake with adaptive, dynamic forms that change with each answer
- LLM symptom interpretation that classifies urgency and the probable department
- Doctor dashboard with patient history, flags and exportable reports
- Subscription model so several clinics can run on one platform
Highlights
- Doctors receive a ranked patient summary rather than a raw intake form
- Multi-clinic subscription setup on a single codebase
Technology
- Next.js
- OpenAI
- LangChain
- Tailwind CSS
- MongoDB
Telehealth intake usually arrives as free text, and someone has to read it before the right doctor sees it. We put an LLM in that first step: it reads the patient’s description, asks the questions a triage nurse would ask next, and sorts the case by urgency and department.
Doctors open a dashboard where cases are already ranked, with the patient’s history and any flags alongside, and can export a report in one click. The platform runs on a subscription model, so additional clinics join the same system instead of getting a separate install.
Client details are kept confidential. On a call we can walk you through this project in more depth, under NDA.