Pylonix - Advanced AI-powered Virtual Health Assistant

Overview

XHALO360 is an advanced AI-powered Virtual Health Assistant (VHA) designed to bridge the gap between patients and healthcare providers. It enables users to securely share health data, receive personalized medical insights, and manage clinical tasks through a natural, conversational interface.

Problem Statement

Patients often struggle with fragmented health data and inefficient communication with providers. Accessing comprehensive medical records such as medications, lab results, and SOAP notes usually requires navigating complex portals, leading to poor health engagement and delayed clinical actions.

Solution

Developed a unified, multi-modal AI chatbot platform that ingests complete patient demographics, clinical records, and real-time vitals to serve as a context-aware health companion. The system automates routine clinical workflows like appointment scheduling and health monitoring while providing instant, layman-friendly explanations of complex medical data.

Key Features

• Context-Aware Medical Intelligence: Integrates patient demographics, medications, conditions, allergies, and historical SOAP notes to provide personalized health guidance.
• Smart Appointment Management: Fully automated booking, rescheduling, and cancellation of appointments via OpenEMR integration.
• Automated Clinical Summarization: Generates daily, weekly, and monthly health summaries and clinical SOAP notes from user interactions.
• Multi-Modal Interaction: Supports text-to-speech and speech-to-text for natural voice conversations in over 50 languages.
• AI-Driven Health Insights: Analyzes vitals and life scores to provide proactive "nudges" and wellness recommendations.

Tech Stack

My Role

Designed the AI architecture, including the development of comprehensive backend APIs for patient and provider portals. I implemented the RAG (Retrieval-Augmented Generation) logic that feeds clinical summaries into Azure OpenAI assistants(vha ChatBOT) to ensure medical accuracy.

Challenges & Solutions

• Challenge: Ensuring real-time synchronization between the AI assistant and the Open-EMR(Electronic Medical Record) system.
• Solution: Built a custom data-sync middleware to map OpenEMR API responses directly into GPT-4o system prompts for instant context updates.

Results & Impact

• Improved patient engagement
• Reduced administrative workload
• Simplified medical understanding

Screenshots

AI Dashboard Interface

AI Dashboard Interface

AI Dashboard Interface

Response Interface

×