Automated 2D Medical Image Matching and Annotation Transfer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Medical imaging exams often present challenges in comparing images due to differences in image characteristics such as scan planes, anatomical locations, and image rotations, leading to difficulties in distinguishing real changes from artifactual changes, which can result in inaccurate assessments.
Innovation Solution
A system that automatically matches and registers medical imaging exams by generating new 2D images through 3D registration and multiplanar reformation of volumetric data, allowing for accurate comparison by aligning image angles and locations, and provides interactive user interfaces for efficient annotation and annotation transfer between exams.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If medical images from different exams are compared directly without matching, then the comparison process is simple and fast, but the accuracy of assessment deteriorates due to artifactual differences in image characteristics
Solution Approach 1:
The system performs preliminary matching and registration of medical images from different exams before comparison. By pre-aligning images based on anatomical landmarks and image characteristics, the system eliminates artifactual differences in scan planes, rotations, and anatomical locations before the physician performs assessment, thereby improving measurement precision without requiring complex manual adjustment during the comparison process
Solution Approach 2:
The system introduces an intermediary matching and registration module that acts as a mediator between raw medical images and the comparison process. This intermediary component automatically adjusts image characteristics (scan planes, rotations, anatomical locations) to create standardized, comparable images, resolving the contradiction by adding a specialized component rather than increasing overall system complexity
2Productivity
If manual annotation is performed on each medical image separately, then annotation precision can be maintained, but the time and effort required increases significantly
Solution Approach 1:
The system automatically copies annotations from reference images to matched images after registration. Once annotations are manually created on one image, the system uses the established matching and registration transformations to automatically transfer these annotations to corresponding locations in other matched images, dramatically improving productivity while maintaining precision through the use of the already-established geometric transformations
Solution Approach 2:
The system merges the annotation process across multiple matched images by allowing annotations to be created once and automatically propagated to all matched images. This combining of annotation operations across multiple images simultaneously reduces the total time and effort required while maintaining consistent annotation precision across the entire set of matched images
Data Source
AI summary
A system is disclosed by which medical imaging exams may be matched and/or registered so as to reduce and/or substantially eliminate artifactual differences between 2D images of the exams. The system may automatically create new, matching 2D images such that two or more exams may be accurately and efficiently compared by a reading physician. The new, matching 2D images may be generated by automated 3D registration of the exams and/or multiplanar reformation of 3D volumetric data acquired during acquisition of one or both exams (e.g., during imaging scans). Rules may be used to automatically determine exams to be compared, and which exams to match to another. Additionally, the system may automatically add indications to a later acquired image that indicate a corresponding earlier acquired image includes annotations. Additionally, the system may allow the physician to easily add and modify similar annotations to the later acquired image by selection of the indications.