Heal AI

From finding your doctor to getting fully healed

Comment

Inspiration

HealAI was inspired by the complexity and challenges individuals face in managing their healthcare journeys. From finding a doctor to analyzing medical reports, the process can be overwhelming and time-consuming. We saw an opportunity to harness the power of AI to simplify and automate this experience. With the integration of AI and Blockchain technology, HealAI empowers users by providing personalized support at every step—from booking appointments to analyzing prescriptions and making recommendations for lifestyle changes. Our vision is to create a holistic health management system, where AI agents handle the heavy lifting, allowing users to focus on their well-being.

What it does

HealAI is an advanced healthcare assistant application that utilizes AI agents to manage the end-to-end health journey. The app features three specialized AI agents:

  • User Agent: Analyzes the user's health condition and suggests doctors for booking appointments. It also manages appointment scheduling and can add reminders to the user's Google Calendar. Additionally, users can book an Uber ride using a custom deep link.

  • Doctor Agent: Receives appointment details and handles the scheduling on the doctor's end. It ensures that both the user and doctor are kept informed throughout the process.

  • Healer Agent: Acts as a medical AI assistant, providing insights from prescriptions and lab reports. It helps users track their medication dosage, reminds them when to take medicine, and offers health insights based on uploaded documents.

HealAI also features location-based doctor recommendations and handles payments securely through the Fetch Blockchain using the Almanac smart contract.

How we built it

We built the backend of HealAI using FastAPI and containerized it as a Docker image for easy deployment (govardhan06/healai:v3). The AI agents were developed using Meta Llama 3.1 70B model for processing health-related prompts, while the frontend was developed using Flutter to ensure a seamless and intuitive user experience.

The Fetch.ai framework was leveraged for creating AI agents that communicate efficiently between users, doctors, and healers. Payments are processed through the Fetch Blockchain, ensuring a secure and automated transaction experience using FET tokens.

Image detection and text analysis was done using a python package called Pytesseract, an open source OCR Engine for optical character recognition for the seamless scanning and analysis of both handwritten and digital medical diagnosis reports.

Challenges we ran into

One of the major challenges was integrating the AI models to process medical data accurately and suggest appropriate actions based on user inputs. We also faced hurdles in ensuring secure and smooth communication between the AI agents, especially for scheduling appointments and processing payments. Another challenge was managing the privacy and security of sensitive health data in a decentralized environment.

Accomplishments that we're proud of

We are proud to have created a robust healthcare system that leverages AI to provide meaningful medical insights and facilitate doctor appointments. Successfully integrating Blockchain technology for secure, automated payments is another achievement we take pride in. Moreover, building a user-friendly interface that simplifies complex healthcare tasks for users is something we’re particularly excited about.

What we learned

Developing HealAI taught us the immense potential of combining AI with Blockchain for solving real-world problems. We learned how to design specialized AI agents to handle specific healthcare tasks and the importance of ensuring privacy and security in a decentralized system. We also gained insights into how AI can assist in medical analysis and provide valuable recommendations for users.

What's next for HealAI

Moving forward, we plan to enhance HealAI by improving the capabilities of the Healer Agent to provide more advanced insights from medical data. We plan to give timely reminders to take medicines to the user through notification from the scanned medical report. We also plan to recommend the user the recommended lifestyle changes for his/her illness to be healed faster. We envision a future where finding a doctor and getting a proper diagnosis is as easy as texting someone on WhatsApp.

Built With

Share this project:

Updates