DICOM De-identification Hashing Collision Resolution

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing DICOM de-identification systems face challenges in efficiently anonymizing large batches of medical records while minimizing collisions among processed records, which can lead to confusion in identifying the origin of medical data.

Innovation Solution

The proposed solution involves applying hashing algorithms and salt values to identification information in DICOM records, ensuring that identifying information is anonymized without compromising data integrity or increasing computing resources needed for anonymization.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional de-identification methods are used on large batches of DICOM records, then processing throughput is reduced, but data privacy protection is maintained

Engineering Contradiction:

Improveprocessing throughputVSAvoiddata privacy protection

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system performs preliminary analysis of the DICOM batch to identify collision-prone identifying information before applying de-identification. This advance preparation allows the system to optimize the de-identification process by pre-determining which fields require anonymization and planning the hashing strategy, thereby improving throughput without compromising privacy protection

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts hashing parameters such as salt values and hash algorithm selection based on the characteristics of the identifying information detected in the batch. By changing these parameters adaptively, the system maintains strong privacy protection while optimizing processing efficiency for different types of medical records

Inventive Principle:

Principle #35Parameter changes

2Reliability

If de-identification is applied to DICOM records, then patient privacy is protected, but collisions between anonymized records may occur leading to loss of data traceability

Engineering Contradiction:

Improvepatient privacy protectionVSAvoiddata traceability

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The system introduces a collision detection and resolution mechanism that acts as an intermediary between the hashing process and the final anonymized output. This intermediary layer monitors for collisions during de-identification and applies resolution strategies such as adding unique suffixes or adjusting hash parameters, thereby maintaining both privacy protection and data traceability

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements feedback loops that continuously monitor the de-identification process for collision patterns. When collisions are detected, the system adjusts its hashing strategy in real-time by modifying salt values or selecting alternative hash algorithms, ensuring that privacy is maintained while minimizing information loss and preserving data traceability

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20250157631A1Systems, methods, and devices for enhanced dicom de-identification

Publication Date: 2025.05.15 HYLAND SOFTWARE INC

AI summary

Techniques, described herein, enable enhanced de-identification solutions for digital imaging and communications in medicine (DICOM) systems and records. DICOM header tags may be de-identified (e.g., anonymized) using one or more hashing algorithms and salt values. Solutions for evaluating and resolving collisions between de-identified. DICOM records may be de-identified in large batches or in smaller batches on a per-request basis. Additional features of the techniques are described herein.