1. Home
  2. Docs
  3. AxiPHYL Health Assistants
  4. AxiPHYL – High-level Architecture

AxiPHYL – High-level Architecture

The AxiPHYL ( AxiPHYL Infinity : Multi Model Primary care AI System) is responsible for hosting Large
Foundation Models, training and inference on large datasets of medical data, speech, video, images and
text. This allows the AxiPHYL Edge Assistant to learn a wide range of skills, such as voice recognition,
natural language understanding, and question answering. The cloud also stores and manages data, such
as user preferences and device information. This data is used to personalize the AxiPHYL Edge Assistant
experience and improve the accuracy of its responses.
The edge (Mobile, Tablet or Kiosk) is responsible for real-time tasks, such as face, voice recognition and
response. This requires low latency, so the edge is typically located close to the user, it can be a mobile or
mobile connected to a local server? Or even connected to Large Models on Cloud and based on latency.
The edge also has increased privacy and security, as it does store sensitive data of only local visitors and
NOT others data , but even this data is highly protected.
When a user interacts with the AxiPHYL Edge Assistant deployed, their input is sent to the cloud if the
connectivity is not there the AxiPHYL Edge Assistant should be intelligent enough to give basic assistance.
Each AxiPHYL Edge Assistance will have clinic specific, town specific cache and few other parameters. The
complete intelligence will get built daily and all three stages of AI should get intelligent day by day. The
cloud then processes the inputs of wide range and generates a response in new markets. The response is
then sent back to the user, who can then interact with the AxiPHYL Edge Assistant again.