Patients on mobile data
- Low-data video and audio-only visits
- The SDK loads only once the doctor admits the patient, so the first screens open fast
- Links work in the phone’s browser
Clone a working consult app for web and Flutter. Patients consent, wait and join from a link; doctors admit them, consult on any connection and sign an AI-drafted note. Built on MediaSFU rooms, recording and AI notes.
The clinic desk books a visit and sends the patient a link by SMS, WhatsApp or email. No app needed; a Flutter app is included if you want one.
The patient tests camera and microphone, then agrees to the consult and, separately, chooses whether it may be recorded.
The patient waits with a live place in the queue. Room access is issued only when the doctor admits them.
Low-data video by default, audio only when the signal is weak, and live translation when doctor and patient speak different languages.
An AI-drafted visit summary and prescription draft arrive for the doctor to edit and sign with their registration number.
Versioned consent text, a separate recording choice, caregiver details and a timestamp, stored with each visit.
When a patient declines, the recording control is replaced by a notice. A live banner tells everyone when recording is on.
The queue runs on your server, so a patient holds no room credentials until the doctor lets them in.
320×180 video by default with an audio-only option, for patients on 3G or prepaid data.
Rooms prefer our Mumbai media servers when you choose the Asia region.
A ready prompt turns the consult transcript into a structured note and prescription draft. Nothing is final until a doctor signs.
Doctor and patient each hear the other in their own language, with your choice of speech and language providers.
Browsers and phones receive a per-room secret after admission, never your API key.
Final recordings are delivered to your own S3-compatible bucket, in the region you pick.
A 20-minute video consult between one doctor and one patient is two streams for 20 minutes. At our published rates:
320×180, the kit’s default for weak connections.
640×360 for patients on good connections.
20 minutes of recorded output at 640×360.
AI notes and translation use the speech and language providers you choose, billed by them directly. Volume discounts and every rate are on the pricing page.
The kit gives you the building blocks; your clinical and legal advisers decide how they apply to your practice. See data residency and compliance.
A small Node server (visits, queue, consent records, audit trail, MediaSFU room calls), a React web app for the clinic desk and the patient, a Flutter patient app, and the AI scribe prompt with its JSON schema. It runs on MediaSFU rooms, recording and AI notes, and you can change any part of it.
Visit and consent records live on your server. Live media runs on MediaSFU media servers in the region the room is placed. Final recordings go to your own bucket and leave MediaSFU storage after 72 hours. The data residency page lists every data type.
MediaSFU does not offer a Business Associate Agreement today, so it is not the right choice for US patient data yet. If you serve US patients, talk to us before going live. Most teams using the kit serve patients in India, Africa and Southeast Asia, where local law applies.
ABDM certification applies to the software that creates health records, which is your platform. The kit gives you a structured, doctor-signed visit note that your ABDM-integrated system can turn into a health record.
The one you choose. The scribe runs on MediaSFU AI notes with your own language-model key or ours; the provider bills its usage directly, with no MediaSFU markup.
Start free, point the kit at your account, and send a visit link to a colleague. We will help you set up recording storage and the AI scribe.