Cryptographic Hash Linking for De-Identified Medical Studies
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for sharing medical imaging data over networks face challenges in linking de-identified data from different studies, as removing identifiers makes it impossible to match studies, and maintaining identifying information compromises patient privacy and security.
Innovation Solution
A service hosted within the organization generates cryptographic hashes from identifying fields and sends them to a cloud-based service for de-identified data, allowing matching of related studies without exposing identifying information, with a cache for improved performance and configuration changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If identifying information is removed from medical studies to protect patient privacy, then patient security and privacy are improved, but the ability to link and match related studies across different organizations is lost
Solution Approach 1:
The patent extracts only the essential identifying fields (patient ID, accession number, study date) from the full medical study data, creates cryptographic hashes of these extracted fields, and stores the hashes in a de-identified manner. This allows linkage capability to be preserved through the hash values while the original identifying information remains protected and separate from the shared cloud storage.
Solution Approach 2:
The patent introduces cryptographic hashes as an intermediary between the original identifying information and the de-identified study data. The hashes serve as a mediator that enables study matching and linkage across organizations without exposing the actual patient identifiers. The hash function acts as a one-way transformation that preserves linkage capability while preventing reverse engineering of patient identities.
2Loss of information
If identifying information is maintained in cloud storage to enable study matching, then study linkage capability is improved, but patient privacy and security are compromised
Solution Approach 1:
The patent extracts only the minimum necessary identifying fields from the complete medical study data. By selecting and extracting only patient ID, accession number, and study date, the system maintains just enough information for study matching while removing all other potentially sensitive patient information from the cloud storage environment.
Solution Approach 2:
The patent uses cryptographic hashes as temporary, disposable intermediaries for study matching. The hashes are generated, used for matching purposes, and then the original identifying information can be securely deleted or kept separate. The hash values themselves serve as short-lived proxies that enable linkage without requiring long-term retention of sensitive patient identifiers in the cloud.
3Measurement precision
If cryptographic hashes are recalculated for all studies whenever configuration changes, then data linkage accuracy is improved, but system performance and processing time deteriorate
Solution Approach 1:
The patent performs preliminary actions by pre-calculating and storing cryptographic hashes for all existing studies during initial data ingestion. When configuration changes occur (such as adding or removing identifying fields), the system performs incremental updates by identifying only the affected studies and recalculating hashes only for those specific studies, rather than reprocessing the entire dataset.
Solution Approach 2:
The patent applies partial action by recalculating cryptographic hashes only for the subset of studies that are affected by configuration changes, rather than recalculating hashes for all studies in the system. The system identifies which studies contain the modified identifying fields and processes only those, reducing unnecessary computational overhead while maintaining matching accuracy for all studies.
Data Source
AI summary
A protected health information (PHI) service is provided which de-identifies medical study data and allows medical providers to control PHI data, and uploads the de-identified data to a remote service system. The PHI service, or an associated service, hosted within an organization performing scans, maintains a database of personally identifiable information, and may generate a cryptographic hash using one or more identifying fields which can then be sent to a service hosting the de-identified data. A cryptographic hash may be further secured by combining the identifying fields with a unique cryptographic key before hashing the value. Matching hashes indicate related studies within the remote service. The service may re-generate and send the cryptographic hashes for each study whenever the configured set of fields changes. This allows all data, historical and new, to be properly linked and related.