Digital Pathology Patient Identity Verification via Image Pattern Synchronization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Human errors are likely to occur when adding a manual barcode in the preprocessing steps of producing a digital slide image in digital pathology systems, such as grossing, waxing, paraffin blocking, or staining, leading to inaccuracies in patient case identity determination.
Innovation Solution
A patient case identity determination method that involves acquiring a digital slide image, requesting and receiving patient/case information from a laboratory information system, comparatively analyzing the image patterns to calculate a synchronization rate, and providing information on the identity match based on this rate, which includes extracting main areas, aligning and classifying sample types, and measuring similarity through shape, color, and edge directivity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual barcode is added in each preprocessing step, then patient case identity can be tracked, but human errors are likely to occur leading to inaccuracies
Solution Approach 1:
The patent replaces the manual mechanical process of adding barcodes with an automated image recognition system. The system captures images from preprocessing steps and automatically extracts identification information through pattern recognition algorithms, eliminating manual intervention and reducing human errors in patient case identity determination.
Solution Approach 2:
The system enables self-service by allowing the digital pathology system to automatically determine patient case identity without external manual input. The image recognition system autonomously processes images from preprocessing steps, extracts identification data, and matches it with patient information, making the system self-sufficient in tracking patient cases.
2Measurement precision
If automated image recognition is implemented, then human errors are reduced, but system complexity increases
Solution Approach 1:
The patent segments the image recognition process into distinct modules: image acquisition from preprocessing steps, pattern extraction, identification information generation, and matching with patient data. This segmentation allows each module to be optimized independently and simplifies the overall system architecture, reducing complexity while maintaining high measurement precision.
Solution Approach 2:
The image recognition system is designed with multi-functionality to handle various types of images from different preprocessing steps (grossing, waxing, paraffin blocking, sectioning, staining). The system uses universal pattern recognition algorithms that can extract identification information from diverse image sources, reducing the need for multiple specialized systems and thereby lowering overall complexity.
Data Source
AI summary
Disclosed is a patient case identity determination method for a digital pathology system, which is performed by a patient case identity determination unit. The patient case identity determination method includes acquiring a digital slide image from a scanner; requesting a laboratory information system (LIS) to send patient/case information including a preprocessing step image that is obtained in a digital slide preprocessing step associated with the digital slide image; receiving the patient/case information including the preprocessing step image from the LIS; comparatively analyzing a pattern of the digital slide image and the preprocessing step image to calculate a synchronization rate and stores the calculated synchronization rate; and providing information about whether the digital slide image is identical to the patient/case information based on the calculated synchronization rate when a client application provides a reading of the digital slide image.