Adaptive Bounding Box Segmentation for Medical Image Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual segmentation of anatomical structures in medical imaging data is time-consuming and prone to user fatigue, especially with increasing scanner resolution, affecting the quality of medical diagnosis and service.
Innovation Solution
An image processing method and system that uses a combination of image processing filters based on partial differential equations, with adaptive models and a 'one mouse click' approach to segment structures like tumors and kidneys, reducing processing time by applying preprocessing and pre-segmentation filters within a bounding shape defined by a seed point.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual segmentation tools are used to identify anatomical structures, then segmentation accuracy can be maintained, but segmentation time increases substantially
Solution Approach 1:
The patent applies preliminary action by performing preprocessing and pre-segmentation operations before the main segmentation task. The method pre-processes the medical image data to enhance structural features, pre-segments the data to identify potential regions of interest, and then refines these preliminary results through adaptive bounding box/ellipsoid models to achieve final accurate segmentation, thereby reducing the time required for manual intervention while maintaining accuracy
Solution Approach 2:
The patent introduces intermediary computational models (adaptive bounding box and ellipsoid models) that act as mediators between the raw image data and the final segmentation result. These intermediary models automatically generate initial segmentation estimates that guide the subsequent refinement process, reducing the burden on manual segmentation while preserving measurement precision through iterative optimization
2Measurement precision
If preprocessing and pre-segmentation are applied to the entire image data, then segmentation quality improves, but processing time increases
Solution Approach 1:
The patent applies local quality by performing preprocessing and pre-segmentation operations selectively within adaptive bounding boxes and ellipsoids that are generated around regions of interest, rather than processing the entire image data uniformly. This localized approach concentrates computational resources on areas where segmentation is needed while skipping irrelevant regions, thereby improving segmentation quality in target areas without proportionally increasing overall processing time
Solution Approach 2:
The patent divides the image processing task into segmented stages: generating adaptive bounding boxes/ellipsoids around potential structures, applying preprocessing only within these bounded regions, performing pre-segmentation locally, and then refining results. This multi-stage segmentation approach allows quality improvement through targeted processing while maintaining productivity by avoiding unnecessary computation in non-relevant image areas
Data Source
AI summary
A seed point is selected inside a structure that is to be segmented in image data. An adaptive model is defined around the seed point, and a preprocessing filter is applied only within the bounding region. A presegmentation of the preprocessed result is performed, and the bounding region is expanded if necessary to accommodate the presegmentation result. An adaptive model for post-processing may be used. The model is translated, rotated and scaled to find a best fit with the pre-segmented data.