AI Cloud-Based Telehealth Framework

No. Ambient documentation is designed as a drafting aid, not as an autonomous charting system. The framework uses LLM-assisted processing of the visit conversation to produce a structured note, and that draft is handed to the clinician to review and edit. Nothing is finalized until the clinician signs it, so the clinician stays the author of record and carries the same clinical and legal responsibility they would for a note they typed themselves. This is deliberate. Language models can misread context, mishear a medication name or leave out a detail, and a review step keeps a person accountable for what enters the patient's record. It also tends to build clinician trust, because the draft is a starting point they control rather than a black box writing into the chart. In practice the clinician reads the note, corrects anything inaccurate, adds the clinical judgment the conversation did not capture, and signs. Once it is signed, the note and visit record are pushed to the EHR through the framework's EHR sync step. How drafts are presented and what review is required can be configured to match the way your clinicians already document. If your organization has specific policies on AI-assisted documentation, those can be mapped during the Fit Assessment so the workflow reflects them from the first day of go-live.

Yes. Both the patient app and the clinician app are white-labeled and configurable, so the people using them see an experience under your brand rather than a third-party telehealth product. That matters for patient trust, because patients who recognize your name at the moment of a visit are more likely to complete it, and for clinicians, who work in an interface that reflects how your organization wants care delivered. Configuration covers more than the logo: the look and the workflow of both apps can be adjusted, so the patient intake flow, the way a visit is joined and the clinician's view of the visit can follow your own processes instead of a fixed design. Because the apps sit on top of the same framework, the underlying capabilities stay in place whatever the branding. That includes AI-assisted triage, WebRTC video, ambient documentation and SMART on FHIR scheduling. Your branding and workflow requirements are mapped during the first step of the deployment path, the Fit Assessment, and applied during configuration. That means you are configuring a telehealth product that already exists, not commissioning a custom build, which is a large part of why typical deployments run 6–10 weeks rather than the much longer timeline of building a telehealth product from a video SDK upward.

Scheduling connects to your EHR through SMART on FHIR, the standard way for applications to read and write patient and appointment data in an EHR. In practice this means telehealth visits are scheduled and recorded inside the same calendar and patient record your team already uses for in-person care, rather than living in a separate telehealth silo that staff have to check and reconcile by hand. The framework reads and writes against the EHR, so a visit booked through the patient app appears where your schedulers and clinicians already look, and the note and visit record created after the visit are pushed back to the patient's chart. Authorization uses OAuth 2.0, and the data exchange is aligned to FHIR R4. During deployment, the second step of the path, Configuration and EHR Mapping, is where SMART on FHIR scheduling is connected to your environment and intake flows are configured. The third step, Sandbox Validation, then runs test visits end to end against your EHR sandbox before anything goes live. That lets your team confirm how appointments, patient records and notes behave in your own EHR, and fix mapping issues, before real patients are involved. Exact connection details depend on your EHR, and they are scoped during the Fit Assessment.

AI-assisted triage handles the step between a patient asking for care and a visit being booked. Instead of a fixed form that asks every patient the same questions, the framework asks adaptive intake questions: each next question depends on how the patient answered the previous one. Based on those responses, the visit is routed to the right visit type or care team. The practical benefit is fewer misrouted visits. When a patient is booked into the wrong type of appointment, the usual outcome is that a clinician discovers the mismatch during the visit, and the patient has to be rescheduled or redirected, which wastes clinician time and delays care. Capturing the right information up front, and routing on it, reduces that rework before a clinician is involved. It is worth being clear about what triage is not. It routes the visit and gathers structured intake information; it does not replace clinical judgment, and clinicians still decide how care is delivered. The intake questionnaires and routing logic are configurable, so the visit types and care teams in your organization are reflected rather than a generic set. Your visit types, triage logic and branding requirements are mapped during the Fit Assessment, and the intake flows are configured and then tested during sandbox validation.

Yes. The framework runs on HIPAA-compliant cloud infrastructure with BAA support, and that applies to the video visit layer as well as the rest of the platform. There are two deployment options. You can deploy the framework into your own cloud environment, on AWS, Azure or GCP, which suits organizations that need data to stay inside infrastructure they already govern. Or Emorphis Health can host it for you, which suits teams that would rather not operate the environment themselves. In both cases the BAA-ready foundation is in place from the start, so your compliance and security reviewers have a defined hosting model to assess rather than a patchwork to assemble. The video itself is cloud-native WebRTC built for clinical use, including waiting rooms, multi-participant visits and screen sharing, so you do not need to stand up your own media infrastructure. The framework's cloud and compliance posture also includes SOC 2 alongside HIPAA. A compliant platform is one part of a compliant program, though: your organization's own policies, consent processes and clinical workflows still matter. These can be reviewed during the Fit Assessment, and the staged go-live, with monitoring and a rollback plan, gives you a controlled way to bring the product into production.