Code base for front end, backend, and rest of related docs
Demo Presentation for Fall Chainlink Hackathon 2021
Provide Radiologists with a tool to double check the pathology labels they make.
- De-identify patient data and store on the Oasis blockchain with labels
- Provide public access to this ever growing database
- Create re-training loop based on new database entries
- Docker image that takes patient MR data and returns a prediction
- Initial database for MR data (Cancer Imaging Archive and Kaggle 2021 RSNA competition)
- Initial ML training pipeline
- Complete Oasis-Docker Image interactions
- Connect Front-End to Oasis backend
- Store Oasis data on IPFS
- Create chainlink external adapter for (IPFS De-Identified Medical Image CID: Label) key value pairs
- Oasis Parcel set-up (current progress not included for private key leak concerns)
- Update white paper for most accurate architecture
- Complete Oasis file handling architecture