3D Volume of Interest Extraction via Interactive Contour Editing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for extracting regions of interest (ROIs) from medical images do not allow for direct editing of volume of interest (VOI) in three-dimensional images, making it difficult to achieve satisfactory VOI extraction.
Innovation Solution
A method that acquires image data in a sectional plane, determines ROIs in multiple slice images, and generates a VOI based on these ROIs and characteristic image data, allowing for real-time observation and adjustment of the extraction process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a contour outline of ROI is drawn on a two-dimensional image and expanded to a three-dimensional VOI, then the VOI extraction process can be completed, but the user cannot directly edit the VOI on a three-dimensional image or understand the effect of the drawing process
Solution Approach 1:
The patent enables direct manipulation of VOI in three-dimensional space by allowing users to draw, erase, and edit contours on 3D rendered images. This transforms the traditional 2D slice-by-slice editing approach into a 3D interactive process, where users can directly modify the volume of interest in the dimension where it is visually represented, thereby improving ease of operation while maintaining process visibility.
2Extent of automation
If region growing algorithm is used to perform image segmentation in a three-dimensional image, then automatic extraction can be achieved, but real-time results are masked by other tissues making it inconvenient for user observation
Solution Approach 1:
The patent extracts and displays the region of interest from the surrounding tissues by allowing users to define ROIs on 3D rendered images. The system then extracts the corresponding VOI from the volume data and displays it separately, removing the masking effect of other tissues. This enables clear observation of the extraction results while maintaining automated region growing algorithms for the actual segmentation process.
3Manufacturing precision
If multiple ROIs are drawn on different slices to extract a VOI, then the extraction coverage can be improved, but it becomes difficult to observe and adjust the real-time extraction results
Solution Approach 1:
The patent allows users to observe and adjust extraction results in three-dimensional space by displaying the VOI on a 3D rendered image. Instead of requiring users to mentally reconstruct the 3D shape from multiple 2D slices, the system presents the extracted volume in its native 3D dimension, making it easy to observe the overall shape and adjust the extraction by drawing or erasing contours on the 3D rendering. This maintains high extraction accuracy while dramatically improving observability and ease of adjustment.
Data Source
AI summary
The present disclosure relates to a system and method for extracting a region of interest. Image data in a first sectional plane may be acquired. The image data in the first sectional plane may include at least one first slice image and one second slice image. A first region of interest (ROI) in the first slice image may be determined. A second ROI in the second slice image may be determined. A first volume of interest (VOI) may be determined based on the first ROI, the second ROI, and characteristic information of the image data in the first sectional plane.