Inspiration
The inspiration behind MediScan stems from the critical need for quick access to health information during emergencies. First responders and health professionals often face challenges in obtaining vital health data promptly, especially when dealing with unconscious or non-communicative individuals. The goal was to create a seamless and privacy-preserving solution to bridge this gap.
What it does
MediScan is an application that enables first responders and health professionals to access a person's health information by simply taking a picture. The use of privacy-preserving technologies ensures the confidentiality of both health and identity information
How we built it
We built a backend API that runs image recognition AI on the images, but only stores the image hashes and not the actual image, so no privacy leakage there. When the first responder sends a picture to the app, it checks if it recognizes the face, and finds the correct hash, and then we use that hash in the Aleo smart contract to verify that the health information belongs to a given person. We use the Aleo blockchain smart contract to verify the first responder's status and only allow access to health information to verified personnel. We use the Aleo Swift SDK on the frontend.
Challenges we ran into
The JavaScript SDK didn't work really well so integrating with Aleo was a challenge. We couldn't manage to get the balance of an address for example. The testnet token faucet was also malfunctioning, which prevented us from actually deploying our contracts to the Aleo testnet.
Accomplishments that we're proud of
Developing our first Aleo smart contract, trying out the mobile and JS SDKs, building a server backend for face recognition, and utilizing AI. Most importantly though, we are proud to solve an actual real-world problem that could be the difference between life and death for many.
What we learned
We learned about building smart contracts in Leo, interacting with the Aleo blockchain, and utilizing face recognition AI.
What's next for MediScan
Looking ahead, we aim to address deployment issues and refine the application for real-world use. Collaborative efforts within the community can contribute to the resolution of challenges, and we aspire to further enhance MediScan's capabilities in ensuring secure and private health information access.
Built With
- ai
- aleo
- face-recognition
- javascript
- leo
- python
- swift
- zk
Log in or sign up for Devpost to join the conversation.