The problem
What was broken before AI
Telehealth clinicians had to listen, reason, document, and keep the patient conversation moving at the same time. SOAP notes and patient instructions are important, but producing them after every visit adds cognitive load, creates after-visit paperwork, and makes the EHR feel like part of the visit rather than a support system.
What changed
What the use case made possible
AI made it possible to capture the visit audio, transcribe it, generate structured draft documentation in parallel sections, and return a SOAP note plus patient instructions quickly enough to fit into the clinician’s existing telehealth workflow.
Why this matters
Why this use case is worth studying
The most useful detail is the modular design. Instead of asking one model call to produce an entire medical note, the team broke the job into narrower documentation tasks: chief complaint and HPI, recent encounters and vitals, assessment and plan, patient instructions, and verification. That gives the system clearer boundaries, makes hallucination controls more explicit, and creates a better review surface for the clinician.
Use this when
When this pattern applies
Use this when a team spends significant time turning conversations into structured, reviewable records and the output must follow a strict format with human approval.
