Visual Media De-identification via Averaged Image Templates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
De-identification of visual media data, particularly in healthcare, is challenging due to the difficulty in identifying and obfuscating sensitive information embedded within the content, especially in diverse medical image formats, which is time-consuming and costly using conventional methods.
Innovation Solution
A visual media de-identification system that merges sequences of images into an averaged image, automatically identifies fixed portions, generates a template with corresponding positions, and obfuscates content within these portions, allowing for semi-automatic refinement and application across various image sequences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional methods are used to de-identify visual media data, then de-identification can be performed, but the process is time-consuming and costly
Solution Approach 1:
The system performs preliminary actions by merging multiple images into an averaged image before de-identification, allowing the identification of fixed portions to be done once and applied across all images in the sequence, thereby reducing the time required for de-identification of each individual image
Solution Approach 2:
The averaged image serves multiple functions: it acts as both a computational tool for identifying fixed portions and as a template for de-identification. The template generated from the averaged image can be applied across multiple image sequences, making the system efficient and reusable
2Extent of automation
If manual methods are used to identify and obfuscate sensitive information, then accuracy can be maintained, but manual effort and costs increase
Solution Approach 1:
The system performs self-service by automatically merging images, identifying fixed portions, generating templates, and applying obfuscation without requiring manual intervention. The process is fully automated from image input to de-identified output, eliminating the need for manual analysis and processing
Solution Approach 2:
The patent replaces manual mechanical processes of image analysis and information identification with automated computational methods. The system uses image processing algorithms and template matching to automatically locate and obfuscate sensitive information, substituting human operators with automated systems
3Manufacturing precision
If de-identification is performed on each image individually, then precision can be maintained, but the process becomes complex and time-consuming
Solution Approach 1:
The system merges multiple individual image processing operations into a single averaged image that captures the common fixed portions. This consolidation allows the de-identification process to be applied once to the averaged image and then replicated across all original images, reducing processing complexity while maintaining precision
Data Source
AI summary
A visual media de-identification system is described. The system includes an image merger and a de-identifying engine. The image merger is configured to merge a sequence of images from a set of visual media data into an averaged image. The de-identifying engine is configured to: bound portions of the averaged image that are determined to be relatively fixed, wherein each bounded portion is identified by a corresponding position in the averaged image; generate a template comprising the bounded portions and the corresponding position for each bounded portion in the averaged image; and de-identify the sequence of images by obfuscating content in the bounded portions.