🤖 Rural Telehealth AI Pre-Screening Simulator
Rural telehealth AI pre-screening simulator for identifying patients in need of remote medical consultation.
Reaching Patients Where Care Is Scarce
Rural clinics sit hours away from many patients.
- 60M+: Rural Americans (live far from specialty care)
- 34 mi: Avg specialist drive (one-way rural travel distance)
- 150+: Rural hospital closures (since 2010 nationwide)
- 38%: Telehealth share of visits (now conducted virtually)
Why rural access lags
Specialists cluster in cities, leaving rural counties underserved.
Some rural counties share a single primary-care doctor.
The telehealth front door
A phone or tablet becomes the entry point to care.
AI as the first triage layer
Software gathers context before any clinician gets involved.
Pre-Screening Questionnaire & Vitals Capture
The AI asks guided questions to build a case file.
- 12: Avg questionnaire length (adaptive symptom questions)
- ~40%: Home vital devices used (of rural patients own one)
- 3 min: Completion time (typical patient intake)
- 8: Symptom fields captured (structured data points)
Adaptive questioning
Each answer shapes which question the AI asks next.
Branching logic trims a long form down to essentials.
Vitals when available
Home pulse oximeters and cuffs feed data automatically.
Building the case file
Symptoms and vitals compile into one structured summary.
Bandwidth & Connectivity Assessment
Rural broadband gaps decide which visit format actually works.
- ~72%: Rural broadband access (below urban coverage rates)
- 500 kbps: Min video bitrate needed (for stable video call)
- ~40 kbps: Audio-only fallback works at (minimal signal required)
- 3×: Dropped-call rate, rural (higher than urban areas)
Measuring the real connection
The system pings for latency, jitter, and available bandwidth.
A weak signal quietly rules out video before it starts.
Signal bars as a decision input
Bandwidth becomes a hard input to the routing logic.
Planning around outages
Cellular relays and satellite links are checked as backup.
Modality Recommendation Engine
Urgency and bandwidth together decide the visit format.
- Urgency ≥3: Video call threshold (and good bandwidth)
- Weak signal: Audio fallback trigger (regardless of urgency)
- Urgent case: In-person referral trigger (overrides connection quality)
- >90%: Modality accuracy target (matched to clinician review)
Three possible pathways
Video, audio-only, and in-person referral are the outcomes.
Severe symptoms always route to in-person, signal or not.
Balancing urgency against bandwidth
Low bandwidth downgrades video visits to audio-only calls.
A transparent recommendation
Clinicians see the reasoning behind each suggested modality.
Consultation Prioritized & Queued
The finished case is ticketed and ranked for review.
- 5: Priority tiers (from Routine to STAT)
- <2 min: Avg time to clinician view (for high-priority cases)
- 100s: Cases auto-sorted daily (across a rural network)
- −45%: Missed urgent cases (with AI-assisted triage)
Issuing the ticket
A structured ticket carries symptoms, vitals, and modality.
Nothing reaches the clinician without its priority label attached.
Sorting the queue
Urgent tickets jump ahead of routine, lower-priority cases.
Handoff to the clinician
The care team opens a pre-screened, ready-to-review case.
Rural telehealth AI pre-screening simulator for identifying patients in need of remote medical consultation.
2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install