Interactive 3D ROI Selection in Medical Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for selecting a region of interest in medical images are cumbersome and do not provide clear visual inspection of three-dimensional regions, making it difficult for users to verify the boundaries of the selected area.
Innovation Solution
A system that includes a user interface for receiving input indicative of a region of interest, a slab selector for automatically determining the position and thickness of a slab based on the region's size, and a visualization subsystem to display the slab, allowing users to verify the boundaries of the region of interest by removing obstructing objects and providing orthogonal views.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the user manually navigates through image slices to place a 3D ROI, then the user can control the region selection, but the process becomes time-consuming and complex
Solution Approach 1:
The system performs preliminary actions by automatically generating a 3D ROI from a 2D ROI selection, then allows the user to review and adjust the pre-generated volume. This reverses the traditional workflow where users manually define each dimension, instead preparing the 3D region in advance for user verification.
Solution Approach 2:
The patent introduces an intermediary automated process that translates simple 2D user input into a comprehensive 3D ROI. This intermediary system handles the complex coordinate transformations and volume calculations, mediating between simple user input and accurate 3D region definition.
2Ease of manufacture
If the depth of the volume of interest is set equal to the thickness of a pre-existing slab, then the setup is simplified, but the user loses freedom to control the region selection
Solution Approach 1:
The system dynamically adjusts the 3D ROI parameters based on user input. The slab thickness and position are not fixed but adapt automatically when the user modifies the 2D ROI, allowing flexible control while maintaining simplified setup. The system recalculates the volume parameters in real-time to match user intentions.
Solution Approach 2:
The patent changes the parameter definition approach: instead of fixing slab thickness as a predetermined value, the system derives slab position and thickness from the 2D ROI characteristics. This parameter transformation allows the same mechanism to serve both simplicity and adaptability.
3Measurement precision
If existing methods are used to define a volume of interest, then the region can be selected, but visual inspection of the three-dimensional region is difficult
Solution Approach 1:
The patent solves the visualization problem by projecting the 3D ROI onto 2D planes and displaying multiple orthogonal views (axial, sagittal, coronal). This dimensionality transformation allows users to inspect the three-dimensional region using two-dimensional displays, making the invisible visible through strategic projection.
Solution Approach 2:
The system segments the 3D ROI visualization into multiple 2D orthogonal projections, allowing users to examine different aspects of the volume separately. By dividing the complex 3D inspection task into multiple 2D views, the system makes comprehensive verification feasible.
4Illumination intensity
If the slab thickness is manually fine-tuned, then the visualization can be optimized, but the process becomes more complex and time-consuming
Solution Approach 1:
The system performs self-service by automatically calculating and setting the optimal slab thickness and position based on the 2D ROI definition. The algorithm autonomously determines the volume parameters without requiring manual intervention, eliminating the need for users to understand or adjust complex thickness parameters.
Solution Approach 2:
The patent replaces manual mechanical adjustment of slab parameters with an automated computational system. Instead of users iteratively tweaking thickness values, the system computationally derives the optimal parameters from the 2D ROI geometry, substituting manual adjustment with algorithmic calculation.
Data Source
Figure 1~2
Figure 3
Figure 4A~4C
AI summary
A system for selecting a region of interest in an image is provided. A user interface (1) is applied for receiving user input indicative of a region of interest (504) in an image (502). A slab selector (2) is provided for selecting a position and a thickness of a slab (503) of the image (502), based on a position and a size of the region of interest (504), the slab (503) comprising at least part of the region of interest (504). A visualization subsystem (3) is provided for visualizing the slab (503). The slab selector (2) is arranged for selecting the position and thickness of the slab (503) such that the slab (503) comprises the region of interest (504) and a thickness of the slab (503) corresponds to a size of the region of interest (504) in a thickness direction of the slab (503). The image comprises a dynamic image. The projection image (505) is obtained by processing voxel values of the dynamic image both along a ray and temporally.