Medical Image Processing Apparatus for Circle of Willis Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional medical image processing techniques struggle to accurately segment the circle of Willis due to its varying morphology and potential occlusions in arterial masks, which limits the effectiveness of template matching methods.
Innovation Solution
A medical image processing apparatus that acquires and divides medical image data into anatomically structured pieces, extracts target regions using characteristic points and interpolation techniques, and outputs information on the extracted regions, enabling accurate segmentation of complex structures like the circle of Willis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If template matching is used to segment the circle of Willis, then the segmentation process is simple and fast, but it fails to accurately extract the circle of Willis when morphological variations or occlusions are present
Solution Approach 1:
The patent segments the circle of Willis extraction process into multiple stages: (1) extracting a large region containing the circle of Willis using template matching, (2) detecting characteristic points within this region, (3) identifying the circle of Willis based on these characteristic points, and (4) extracting the final segmented region. This multi-stage segmentation approach maintains the speed advantage of template matching while improving accuracy through characteristic point verification.
Solution Approach 2:
The patent performs preliminary extraction of a large region containing the circle of Willis using template matching before conducting the more accurate characteristic point detection. This preliminary action narrows down the search area, making subsequent precise extraction more efficient and accurate while maintaining overall processing speed.
2Device complexity
If simple template matching is applied, then the processing method is straightforward, but it cannot handle various shapes and morphological variations of the circle of Willis
Solution Approach 1:
The patent employs dynamic adjustment of extraction criteria based on detected characteristic points. Instead of using a fixed template, the system adapts the extraction region and boundaries according to the actual positions and distributions of characteristic points, enabling it to handle various morphological variations of the circle of Willis.
Solution Approach 2:
The patent changes multiple parameters during the extraction process, including the size and position of the extraction region, the criteria for identifying characteristic points, and the boundaries definition. These parameter adjustments allow the system to adapt to different shapes and morphologies of the circle of Willis while maintaining a relatively simple overall processing framework.
3Device complexity
If template matching is used on occluded arterial masks, then the processing remains simple, but structures beyond occlusions cannot be identified
Solution Approach 1:
The patent segments the processing into an initial template matching stage that can handle occlusions by extracting a larger region, followed by a characteristic point detection stage that identifies structures despite occlusions. This segmentation allows the system to overcome occlusion limitations while maintaining processing simplicity.
Solution Approach 2:
The patent introduces characteristic point detection as an intermediary step between template matching and final extraction. This intermediary process helps identify the circle of Willis structures even when direct template matching fails due to occlusions, thereby improving identification reliability without significantly increasing processing complexity.
Data Source
AI summary
A medical image processing apparatus includes processing circuitry configured to acquire medical image data representing a region including a target site, divide the medical image data into a plurality of pieces of divisional data according to an anatomical structure, extract a target region corresponding to the target site from each of the pieces of divisional data according to a particular condition, and cause an output circuit to output information on the extracted target region.