Patient-centered, privacy protected testing framework leveraging blockchains and distributed data storage to run a point of care or home test. Objectives are to increase safety for both patients and doctors, to reduce the lead time around current processes, to have identity based provenance, consumer-based consent based data share and de-identified reporting and a privacy-protected minimal proof of safety for use in a possible passport. Working with Lab partners. Initial focus is India, but looking to build as neutral as possible. Need identity experts, d-app developers and REST API.
Thanks for the contributions.
-
Presentation: Please follow the link
-
Demo video: Please follow the link
-
FlexiTest-SolutionArchitectureAndDesign: Please follow the link
- The Presentation Demonstration, the Video demonstration and GitHub illustrations vividly
explains the issues experienced in real-world issues we faced
the enterprise systems and processes exposed limitations in availability, staffing and hence timely availability of
results even after testing results were available.
gaps in telehealth systems for full auditability / traceability and non-repudiation as more businesses are relying on
these test results
issues around accountability, transparency, speediness, and privacy protected reporting of de-identified , aggregated
numbers for public health use
and the realization that re-opening of the economy and wide- spread high scale SAFE testing are 2 sides of the same coin.
the opportunities to solve these real world problems
to increase not just patient safety , but doctor / health worker safety
to have a key tool for safe re-opening of the economy with privacy based minimalistic proof needed for overall safety yet protecting civil liberties
2.The solution addresses the primary goal of COVID 19 hackathon to solve the real world problems identified
by leveraging decentralization based technologies including IPFS, the ethereum blockchain to prove existence of key data and metadata across organizations for an audit-ready yet privacy protected designs based on the ethereum stack
by using components already available in the blockchain ecosystem including DID specifications and components,
secure distributed file storage systems
-
The solution addresses the gaps in current systems
by having key cross-organizational patient data/ meta-data secured for accountability, and cryptographically secured data to quickly send information between parties in the teletext workflow reducing delays allowing whichever parties who are covered entities that have the consented data to report to public health without delays de-identifying personal data and also reducing chances of re-identification
-
The solution entails innovative elements to
incorporate standards like FHIR in to the solution for the business objects use Zero Knowledge proof designs based on the baseline protocol to those standard objects , and to design the arithmetic circuits to find if the user has been tested, and the status of their test results to the checker archetype in a minimalistic proof.
Please refer to the details at
https://github.com/chainaim3003/FlexiTest/blob/master/README.md
https://github.com/chainaim3003/FlexiTest/blob/master/zkp/README.md
At Home Collection; At Lab Execution
At Home Collection; At Home Execution ( At Home Antibody Tests )
The Checker is an abstraction and could include individuals or organizations such as airport authorities, event management authorities, law enforcement or any such organization who has gained and can prove the verifier status.