The problem DERMAI solves

The user is greeted by an friendly user interface.
There are two types of users for our platform that is patients who want to get diagnosed and healthcare professionals which include doctors and researchers.
After logging into our platform which is done using magic.link that is a walletless onboarding platform hence the user doent need to already have a wallet linked like metamask.
for patients once they signup they are greeted by a dashboard where they can upload the images of their skin conditions and enter their symptoms as well.
Based on both the input the model will predict what skin condition the patient is experiencing, the model can give diagnosis at an accuracy of 98%
The system employs AI algorithms trained on a vast dataset of dermatological images and medical knowledge to analyze the input data.For image-based diagnosis, the AI algorithms identify patterns, lesions, or abnormalities in the skin images to generate preliminary diagnoses.
For symptom-based diagnosis, the AI algorithms analyze the symptoms provided by the user to generate potential differential diagnoses.
On the same dashboard patients have the option of sharing their diagnosis with research institutes and medical professionals.
To encourage user participation and data sharing, the system rewards users with Ethereum points for sharing their diagnosis and contributing to the platform.
Users can accumulate Ethereum points, which can be used for various cryptocurrency applications or incentives within the platform.
User data, including images, symptoms, and diagnoses, is securely stored using decentralized storage mechanisms such as blockchain technology.
Decentralized storage ensures the safety, transparency, and integrity of user data, protecting it from unauthorized access or tampering.
Smart contracts are employed to enforce the immutability and tamper-proof nature of the data stored on the blockchain.
Once diagnostic data is recorded, it becomes immutable and cannot be altered, providing

Challenges we ran into

We faced difficulties while integrating chat application, we were facing issues during socket connection.
Walletless Onboarding with Magic.link:
Our team was accustomed to custodial wallets, but we needed to explore new methods like wallet-less onboarding to onboard the next wave of users who are more familiar with Web2 social logins. Recent updates to Magic.link added complexity to the transition from dedicated to universal wallet.
Escrow Implementation: Designing a secure escrow system that holds funds until patient consent is obtained.
IPFS Storage Transition:
Our team was proficient with web3.storage for IPFS, but recent changes posed difficulties. Transitioning to Pinata required time to understand differences in features, APIs, and workflows.