Medical Image Association Using Vascular Models and Biometric Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for associating medical images with a patient are hindered by typographic errors, duplicate biographical data, and variations in medical image acquisition protocols and equipment across different institutions, leading to challenges in accurately linking images over time, especially due to changes caused by age, disease progression, and treatment.
Innovation Solution
A computer-implemented system and method that uses patient biographical data and vascular models to determine a matching score between medical images, adjusting for differences in image acquisition and incorporating additional biometric patterns to ensure accurate patient identification across multiple institutions and platforms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If patient biographical data is used for association, then the association process is simple, but typographic errors and duplicate data reduce reliability
Solution Approach 1:
The patent uses biometric data (vein patterns, fingerprint, facial recognition) as an intermediary verification mechanism between the patient biographical data and the medical images. This intermediary layer validates patient identity independently of biographical data quality, resolving the contradiction by maintaining simple association processes while ensuring high reliability through biometric confirmation.
Solution Approach 2:
The patent replaces the mechanical/manual process of verifying patient identity through biographical data entry and comparison with an automated biometric recognition system. This substitution eliminates typographic errors and duplicate data issues by using physiological traits that are difficult to falsify or misrecord, thereby maintaining operational simplicity while dramatically improving identification accuracy.
2Measurement precision
If standardized image acquisition is used, then biometric matching accuracy is improved, but adaptability to different medical institutions and protocols is reduced
Solution Approach 1:
The patent employs multiple biometric modalities (vein patterns, fingerprint, facial recognition) that can be applied universally across different medical institutions and imaging protocols. These biometric methods are not dependent on specific image acquisition parameters, allowing the system to maintain high matching accuracy while adapting to diverse institutional environments and imaging standards.
Solution Approach 2:
The patent changes the fundamental parameters used for patient identification from image acquisition parameters (which vary by institution) to biometric parameters (which are intrinsic to the patient). By using physiological traits such as vein patterns and fingerprint that remain relatively stable regardless of imaging protocol variations, the system achieves both high precision and broad adaptability.
3Reliability
If multiple biometric methods are used, then patient identification reliability is improved, but system complexity increases
Solution Approach 1:
The patent segments the biometric verification process into distinct, independent modules (vein pattern recognition, fingerprint recognition, facial recognition). Each module can be implemented separately and contributes to the overall reliability through modular redundancy. This segmentation allows the system to achieve high identification accuracy while managing complexity through organized, independent components rather than a monolithic complex system.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods are disclosed for associating medical images with a patient. One method includes: receiving two or more medical images of patient anatomy in an electronic storage medium; generating an anatomical model for each of the received medical images; comparing the generated anatomical models; determining a score assessing the likelihood that the two or more medical images belong to the same patient, using the comparison of the generated anatomical models; and outputting the score to an electronic storage medium or display.