Skip to Content
Althea Studio

Widgets

The Widgets layer is a fundamental component of our software stack, consisting of small, task-specific modules that perform essential functions within workflows. Each widget is designed to handle a particular aspect of patient engagement or healthcare operations, enabling our AI agents to execute complex processes efficiently and effectively.

Example widgets

Below is a detailed description of the various widgets our system can handle, along with use cases to illustrate their application.

  1. Intake:

    • Description: This widget is responsible for collecting initial data from patients. It ensures that all necessary information is gathered at the beginning of an interaction.
    • Use Case: When a patient calls to schedule an appointment, the Intake widget collects basic information such as the patient's name, contact details, and reason for the appointment.

  2. Authentication:

    • Description: This widget verifies the identity of patients by cross-referencing their information with Electronic Health Records (EHR) or other authentication methods.
    • Use Case: Before providing sensitive information or scheduling an appointment, the AI agent uses the Authentication widget to confirm the patient's identity.

  3. Navigation:

    • Description: This widget guides users through various processes, ensuring they reach the desired outcome efficiently.
    • Use Case: During the appointment scheduling process, the Navigation widget helps the patient navigate through available time slots and select a suitable appointment time.

  4. Information Delivery:

    • Description: This widget provides relevant information to patients, ensuring they are informed and engaged.
    • Use Case: After scheduling an appointment, the Information Delivery widget sends a confirmation message and any necessary pre-appointment instructions to the patient.

  5. Data Validation:

    • Description: This widget ensures the accuracy and completeness of collected data before it is integrated into the healthcare system's databases.
    • Use Case: When updating patient records, the Data Validation widget checks the submitted information for errors or inconsistencies.

  6. Data Integration:

    • Description: This widget updates patient records in the healthcare system's databases with validated information.
    • Use Case: After validating new patient information, the Data Integration widget updates the patient's EHR with the latest data.

  7. Scheduling:

    • Description: This widget manages the scheduling of appointments, including checking availability, negotiating times, and confirming bookings.
    • Use Case: When a patient requests an appointment, the Scheduling widget checks the availability of healthcare providers, negotiates a suitable time, and confirms the booking.

  8. Data Analytics:

    • Description: This widget analyzes patient data to identify trends, gaps, and opportunities for improving patient care.
    • Use Case: The Data Analytics widget identifies patients who have missed their annual screenings and flags them for follow-up.

Custom Widgets

In addition to the standard widgets, users can create custom widgets to meet specific requirements. Custom widgets can be defined using two primary methods: prompt-based widgets and recipe-based widgets.

Prompt-Based Widgets

  • Description: These widgets are defined using a particular prompting scheme to instruct the engine. Users can also define functions to enable their agent to execute commands interactively.
  • Function Definition: Users can define new connections with their data sources and APIs based on requirements. If a function is not present in our function repository, users can create and integrate new functions.
  • Use Case: A healthcare provider needs a custom widget to interactively guide patients through a symptom checker. The provider defines prompts and functions to connect to their symptom database and APIs, enabling the AI agent to provide personalized recommendations.

Recipe-Based Widgets

  • Description: These widgets are useful for conditional situations in a workflow that require fine-grained control, such as triaging or form fill-out. Users can import a JSON file in a specific format, and our AI will handle the rest.
  • Use Case: A healthcare provider wants to create a triage system that asks patients a series of conditional questions to determine the urgency of their condition according to the Schmitt-Thompson protocol. The provider imports a JSON file with the triage logic, and the AI agent uses this recipe-based widget to guide patients through the process.

Controls

For widgets, there are also parameters that can be configured to optimize their performance and behavior. These controls include:

  • Output Format: Define the format in which the widget's output should be presented, ensuring compatibility with other system components.
  • LLM Selection: Users can choose from supported Large Language Models (LLMs) to power their widgets. Supported engines include:
    • LLM Options: Althea, OpenAI, Azure(OpenAI), Anthropic, Google

By leveraging these detailed widgets, custom widgets, and controls, our AI agents can effectively automate and manage various aspects of patient engagement and healthcare operations, ensuring a seamless and efficient experience for both patients and healthcare providers.

previousWorkflows introduction