Automatic Organ Finding via Geometric Shape Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical imaging technologies face challenges in automating the detection of anatomical structures and abnormalities in medical images, particularly in aligning cardiac PET studies due to variability in operator interpretation and image orientation.
Innovation Solution
A system and method for automatically finding an organ in image data by transforming the image to approximate a predetermined view, using normalized cross-correlation to match a synthesized geometric shape, such as a cylinder, to locate the organ in both transformed and original image spaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual methods are used to locate and reorient the heart in cardiac PET studies, then the heart can be positioned for analysis, but there is variability between operators and successive attempts
Solution Approach 1:
The system enables automatic organ finding and reorientation without human intervention. The computer automatically locates the heart in PET images, determines its orientation, and reorients the images to a standard view, eliminating operator variability and achieving consistent, reproducible results across different users and sessions.
Solution Approach 2:
The patent replaces manual mechanical manipulation of images by operators with an automated computer-based system using image processing algorithms. The system automatically transforms images to approximate predetermined views and uses normalized cross-correlation to match synthesized geometric shapes, substituting human manual operations with automated computational methods.
2Reliability
If automated organ finding is implemented, then operator variability is reduced, but the complexity of the system increases
Solution Approach 1:
The system creates synthesized geometric shape copies (such as cylindrical models of the heart) that represent the expected organ geometry. These synthesized templates are then used as references for matching against actual organ images through normalized cross-correlation, simplifying the automated recognition process while maintaining reliability.
Solution Approach 2:
The system transforms images to approximate predetermined views by applying geometric transformations and adjusting imaging parameters. This parameter transformation approach converts complex 3D organ localization into a standardized 2D matching problem, reducing system complexity while improving reliability.
Data Source
AI summary
A framework for automatically finding an organ in image data. In accordance with one aspect, a predetermined view of the organ is approximated by transforming the original image data to generate transformed image data. A best-match region in the transformed image data that best matches a synthesized geometric shape may then be found. The best-match region may be transformed into a volume space of the original image data to generate a location of the organ.