Distributed AI Inference With Privacy-Preserving Capsule Computing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Access to diverse, high-fidelity, privacy-protected data is a significant barrier for developing robust and generalizable artificial intelligence (AI) applications, particularly in healthcare, due to regulatory, legal, and ethical requirements for maintaining patient information privacy, which hinders timely access and sharing of sensitive data across organizations.
Innovation Solution
A system and method for developing AI algorithms using a secure capsule computing framework that integrates with multiple data sources, enabling privacy-preserving data processing and federated training, allowing access to diverse data while maintaining privacy through techniques like differential privacy and homomorphic encryption, ensuring data remains within the control of the data owner.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is shared across organizations to enable diverse data access for AI development, then data diversity and availability are improved, but data privacy and security are compromised
Solution Approach 1:
The patent introduces a privacy-preserving computing framework that acts as an intermediary between data owners and AI developers. This framework enables data to be processed and shared while maintaining privacy through techniques like differential privacy, homomorphic encryption, and secure multi-party computation, thus resolving the contradiction between data availability and privacy protection
Solution Approach 2:
Instead of sharing actual sensitive data, the system creates and shares computational representations or models that capture patterns and insights without exposing the underlying private data. This allows AI development using diverse data while the original data remains protected and localized at data owners' premises
2Ease of operation
If data is centralized for easier access and processing, then ease of operation is improved, but data privacy control and security are worsened
Solution Approach 1:
The system segments data processing operations into distributed components where data remains localized at different organizational premises. Computational tasks are divided and executed securely across multiple locations using encrypted communication and distributed computing frameworks, enabling ease of operation while maintaining data privacy and control
3Reliability
If regulatory compliance requirements are enforced to protect patient privacy, then data security is improved, but data access speed and productivity are reduced
Solution Approach 1:
The system performs privacy-preserving transformations and data preparation actions in advance before actual AI processing occurs. By pre-computing and pre-processing data under privacy constraints, the system establishes compliant data structures and formats ahead of time, enabling faster subsequent access and processing while maintaining regulatory compliance
Data Source
AI summary
The present disclosure relates to techniques for developing artificial intelligence algorithms by distributing analytics to multiple sources of privacy protected, harmonized data. Particularly, aspects are directed to a computer implemented method that includes receiving an algorithm and input data requirements associated with the algorithm, identifying data assets as being available from a data host based on the input data requirements, curating the data assets within a data storage structure that is within infrastructure of the data host, and integrating the algorithm into a secure capsule computing framework. The secure capsule computing framework serves the algorithm to the data assets within the data storage structure in a secure manner that preserves privacy of the data assets and the algorithm. The computer implemented method further includes running the data assets through the algorithm to obtain an inference.