System for Anonymizing and Aggregating Protected Health Information

Abstract

Embodiments of the invention relate to systems for anonymizing and aggregating protected health information. A patient anonymization system includes multiple hashing devices and data sources. Each device receives medical records containing at least confidential protected health information (PHI). An obfuscated value is appended to each secret PHI, and a hash is generated, which replaces the secret PHI to generate an anonymized record. The main patient index server aggregates the anonymized records. The vector and cluster matching engine determines whether the anonymized record matches the unique patient identifier corresponding to the second anonymized record. A comparison vector is generated by comparing the hash value of the confidential PHI to the hash value in the second anonymized record, and cross-multiplied with the confidence vector with match-based weights. This produces a match confidence level, which is compared to a threshold. If the threshold is met, the anonymized record is mapped to a unique patient identifier associated with the second record.

Description

technical field [0001] The present disclosure relates to aggregating patient medical records, and in particular to aggregating and organizing medical records in a manner that protects a patient's identity. Background technique [0002] Patient medical records are increasingly being digitized and stored in computerized databases. Data privacy and security issues and compliance with applicable laws and regulations are thus paramount. For example, in the United States, HIPAA (Health Insurance Portability and Accountability Act) requires that patient medical records be kept confidential and not be released to third parties without authorization. However, it would be advantageous for different entities to have access to certain medical records for research, clinical research, and diagnostic purposes. However, many regulations, including HIPAA, do not allow unrelated or independent entities to aggregate medical records because such aggregation could allow entities to identify in...

AI Technical Summary

Problems solved by technology

However, many regulations, including HIPAA, do not allow the aggregation of medical records by unrelated or independent entities because such aggregation could allow entities to identify individuals associated with the medical records, resulting in a privacy violation

[0003] Additionally, even when medical records are properly obtained, such records may be incomplete, erroneous, and / or ambiguous

Therefore, aggregating and correlating the medical records corresponding to a specific patient is difficult regardless of privacy and compliance concerns

Images