MOVO-X
Loading MOVO-X…
A patient check-in kiosk looks simple from the outside: walk up, tap, get a ticket. The engineering behind those 90 seconds involves ID chip reading, NLP processing, EMR write, and queue routing — all completing before the patient finishes reading their queue number.
Clinicians and administrators evaluating kiosk technology often receive vendor promises without technical depth. Understanding how a check-in kiosk works end-to-end is the prerequisite for evaluating whether it will survive real clinical operations.
The kiosk identifies the patient by NFC chip read (MyKad/passport), QR code scan, or manual IC entry. NFC is the fastest and most accurate — the chip is read in under 2 seconds with no manual input required.
The system checks whether the patient has a scheduled appointment. If yes, the appointment is retrieved and pre-filled. If no, a walk-in intake flow begins. Both paths complete in under 30 seconds.
The patient describes their reason for visit. NLP processes the input in real-time, extracts structured symptoms, and maps to ICD-10/SNOMED-CT codes. This step takes 20–40 seconds.
The AI routing engine assigns the patient to the correct doctor and queue position based on appointment, acuity score, doctor availability, and clinic-configured priority rules. Assignment is instantaneous.
All captured data (demographics, chief complaint, ICD-10 codes, insurance status, consent) is written to the EMR as a FHIR R4 Encounter resource. Write confirms in under 500 ms.
The kiosk prints a 58mm thermal receipt with queue number, estimated wait, doctor name, and room number. Simultaneously, a WhatsApp message is sent with the same information plus a queue-tracking link.
End-to-end check-in in 90 seconds average
EMR receives complete structured data before patient sits in the waiting area
Queue display updates within 300 ms of check-in completion
WhatsApp confirmation delivered within 5 seconds of ticket print
The check-in pipeline is an event-driven sequence: patient identification → appointment lookup → intake capture → acuity scoring → queue assignment → EMR write → receipt print → WhatsApp send. Each step is a discrete microservice communicating over WebSocket and REST. Steps 1–4 happen client-side on the kiosk; steps 5–7 happen server-side in parallel. The kiosk shows the receipt before the server confirmation returns — optimistic UI with server reconciliation within 2 seconds.
The kiosk still issues the queue ticket (optimistic). The failed EMR write is queued locally and retried with exponential back-off. A staff alert is raised if the retry fails after 3 attempts. No patient is ever turned away due to an EMR timeout.
The routing engine evaluates: appointment assignment (highest priority), specialty matching, current doctor queue depth, doctor availability flag, and clinic-configured priority rules. The result is the doctor with the best combination of relevance and availability.
Yes. Steps 1–4 run entirely on-device. The kiosk queues the EMR write and queue assignment locally and syncs when connectivity returns. The thermal receipt is printed from local state.
RK3566 ARM Cortex-A55 quad-core SoC, Android 11, 4GB RAM, 64GB eMMC, 10" IPS touchscreen, USB-C, built-in microphone and speaker, SCR301 NFC reader, 58mm USB thermal printer, front-facing camera.
The patient's phone number is captured during the check-in flow (from ID chip data or manual entry on first visit, then stored for subsequent visits). WhatsApp Business API sends the template message to that number immediately after queue assignment.
Tell us about your operation. We'll send a tailored proposal — architecture, integration scope, deployment timeline, and total investment — within hours.