Skip to Content
Althea Studio

Workflows

The Workflows layer is a crucial component of our software stack, responsible for organizing and executing sequences of widgets to perform specific tasks. Each workflow is designed to handle a particular aspect of patient engagement or healthcare operations, ensuring that our AI agents can carry out complex processes efficiently and effectively.

Example workflows

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

  1. Outreach:

    • Description: This workflow is designed to engage with patients for follow-ups, reminders, or information dissemination. It ensures that patients receive timely and relevant communications.
    • Components:
      • Widgets: Intake, Authentication, Information Delivery
    • Use Case: A healthcare provider wants to remind all diabetic patients about their annual check-up. The AI agent uses the Outreach workflow to contact each patient, verify their identity, provide the reminder, and record their responses.

  2. Gap Closure:

    • Description: This workflow identifies and addresses gaps in patient care, ensuring that patients receive comprehensive and continuous care.
    • Components:
      • Widgets: Data Analytics, Intake, Authentication, Information Delivery, Scheduling
    • Use Case: A healthcare provider identifies that several patients have missed their annual screenings. The AI agent uses the Gap Closure workflow to contact these patients, inform them about the missed screenings, and schedule follow-up appointments.

  3. Scheduling:

    • Description: This workflow manages appointment scheduling, ensuring that patients can book, reschedule, or cancel appointments efficiently.
    • Components:
      • Widgets: Intake, Authentication, Navigation, Information Delivery
    • Use Case: A patient calls to schedule an appointment. The AI agent uses the Scheduling workflow to authenticate the patient, check the availability of healthcare providers, negotiate a suitable time, and confirm the appointment.

  4. Form Intake:

    • Description: This workflow handles the collection and processing of patient forms, ensuring that all necessary information is gathered and validated.
    • Components:
      • Widgets: Intake, Authentication, Data Validation, Data Integration
    • Use Case: A healthcare provider needs to update patient records with new information. The AI agent uses the Form Intake workflow to collect updated information from patients, validate it, and integrate it into the healthcare system's databases.

  5. Information Delivery:

    • Description: This workflow focuses on providing relevant information to patients, ensuring they are informed and engaged.
    • Components:
      • Widgets: Intake, Authentication, Information Delivery
    • Use Case: A patient needs regular updates about their treatment plan. The AI agent uses the Information Delivery workflow to send periodic updates, educational materials, and reminders to the patient, ensuring they stay informed and engaged.

  6. Complex Workflow Execution:

    • Description: This workflow involves combining multiple workflows to achieve complex, multi-step objectives. It allows for the creation of comprehensive processes that address various patient needs.
    • Components:
      • Workflows: Intake, Authentication, Scheduling, Information Delivery
    • Use Case: A healthcare provider wants to conduct a comprehensive health check-up campaign. The AI agent combines the Intake, Authentication, Scheduling, and Information Delivery workflows to first collect patient information, authenticate their identity, schedule appointments, and provide necessary information about the check-up.

Controls

Each workflow includes a set of controls to ensure proper execution and performance. These controls help in assessing the workflow's effectiveness and making necessary adjustments. Key features include:

  • Performance Metrics: Track key performance indicators (KPIs) such as workflow completion rates, response times, and patient satisfaction.
  • Error Handling: Monitor for errors or exceptions during workflow execution and implement corrective actions.
  • Data Analytics: Analyze collected data to generate insights and improve future workflow performance.
  • User Feedback: Collect feedback from patients and healthcare providers to refine and enhance the AI agent's capabilities.

Overarching Parameters

In addition to the specific controls for each workflow, there are overarching parameters that can be configured to optimize the performance and behavior of the AI agents. These parameters include:

  • Number of Utterances: Define the maximum number of utterances allowed in a single interaction to ensure concise and effective communication.
  • Voice-Related Models and Parameters:
    • STT (Speech-to-Text) Engine Selection: Users can choose between Althea's proprietary AI models or well-known supported engines such as Azure, Google, AWS, and Deepgram.
    • TTS (Text-to-Speech) Engine Selection: Users can select from Azure, Google, AWS, Deepgram, 11 Labs, and OpenAI for generating natural-sounding speech.
  • Standard Delay: Set a standard delay between interactions to ensure a natural flow of conversation.
  • Turn Prediction: Implement turn prediction to anticipate user responses and improve the efficiency of interactions.
  • Voice Tagging: Use voice tagging to label and categorize different parts of the conversation for better analysis and understanding.

By leveraging these controls and overarching parameters, users can customize and optimize the performance of the AI agents to meet their specific needs and preferences, ensuring a seamless and efficient experience for both patients and healthcare providers.

previousTasks introduction