Toboggan Object Detection in CT Cutting Planes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical image analysis techniques, such as tobogganing, face challenges in efficiently detecting spherical or ellipsoidal structures like polyps in CT images due to high computational complexity and the need for processing entire image datasets, which is unnecessary given known object locations.
Innovation Solution
A method and system for toboggan-based object detection in cutting planes that determines a region of interest, computes toboggan potential, extracts cutting planes, and performs tobogganing to form clusters, storing elements in cluster-member, outer-border, and inner-border lists to efficiently identify objects, reducing computational complexity and improving detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional tobogganing processes the entire image dataset to determine toboggan clusters, then detection accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent divides the 3D image dataset into multiple 2D cutting planes (axial, coronal, and sagittal views) and processes each plane separately using toboggan algorithm. This segmentation allows the system to focus computational resources on specific 2D slices rather than processing the entire 3D volume, reducing computational complexity while maintaining detection accuracy through multi-planar analysis.
2Measurement precision
If watershed segmentation processes most pixels/voxels in all 2D cutting planes, then object separation is improved, but processing time increases
Solution Approach 1:
The patent applies watershed segmentation selectively only to regions identified as potential objects of interest in the 2D cutting planes, rather than processing all pixels/voxels in the entire dataset. This partial action approach maintains effective object separation where needed while avoiding unnecessary computational overhead in regions without objects, thereby reducing overall processing time.
3Productivity
If fast tobogganing starts from a specified location and forms local clusters, then processing speed is improved, but detection coverage may be reduced
Solution Approach 1:
The patent compensates for the reduced detection coverage of fast tobogganing by analyzing multiple 2D cutting planes (axial, coronal, and sagittal dimensions) from the same 3D dataset. Objects that may be missed in one plane are likely to appear in other planes, ensuring comprehensive detection coverage across all three spatial dimensions while maintaining the processing speed benefits of fast tobogganing.
Data Source
AI summary
A system and method for toboggan-based object detection in cutting planes are provided. A method for detecting an object in an image includes: determining a region of interest (ROI) in the image; determining a toboggan potential for each image element in the ROI; extracting a plurality of cutting planes from the ROI; and performing a tobogganing in the cutting planes to form a toboggan cluster to determine a location of the object, wherein image elements inside the toboggan cluster are stored in a cluster-member list, image elements on an outer-border of the toboggan cluster are stored in an outer-border list and image elements on an inner-border of the toboggan cluster are stored in an inner-border list.