Scan Region Determination Using Spatial Transformation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medical imaging systems face challenges in reliably determining the scan region, especially when localizer images contain multiple objects, leading to inconsistent and inefficient field of view determination.
Innovation Solution
A scan region determining apparatus that uses spatial transformation to register overview and template images, initializing the registration process with a bounding box to reduce the likelihood of local minima traps, allowing for robust and fast determination of the scan region by projecting the template scan region onto the overview image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional field of view determination methods are used on localizer images, then the scan region can be determined, but the determination is unreliable when multiple objects are present and may trap in local minima
Solution Approach 1:
The patent applies preliminary action by first creating a binary mask from the localizer image to identify the body region before performing field of view determination. This preliminary segmentation step separates the body from other objects, ensuring that subsequent field of view calculation is performed only on the relevant anatomical region, thereby preventing local minima traps and improving reliability when multiple objects are present in the image.
2Measurement precision
If manual field of view determination is performed, then accuracy can be maintained, but time consumption increases significantly
Solution Approach 1:
The patent implements self-service by enabling the system to automatically determine the field of view through algorithmic processing of the localizer image. The binary mask generation and connected component analysis automatically identify the body region and calculate the minimal bounding box without requiring manual intervention, thereby maintaining accuracy while dramatically reducing the time required for scan preparation.
3Reliability
If the scan region includes all visible objects in the localizer image, then no objects are missed, but unnecessary radiation exposure increases
Solution Approach 1:
The patent applies the extraction principle by isolating only the body-related objects from the localizer image through binary mask generation and connected component analysis. By extracting the largest connected component representing the patient's body and excluding other objects (such as medical devices, positioning aids, or unrelated anatomy), the system defines a minimal scan region that ensures completeness for diagnostic purposes while minimizing unnecessary radiation exposure to non-target areas.
Data Source
Figure 1
Figure 2~5
Figure 6
AI summary
The invention relates to a scan region determining apparatus (12) for determining a scan region of a subject to be scanned by a scanning system (10) like a computed tomography system. A spatial transformation defining a registration of an overview image and a template image with respect to each other is determined, wherein initially the overview image and the template image are registered by using an element position indicator being indicative of a position of an element of the subject with respect to the overview image. A template scan region is defined with respect to the template image, wherein a final scan region is determined by projecting the template scan region onto the overview image by using the determined spatial transformation. The registration and thus the determination of the spatial transformation are very robust, which improves the quality of determining the final scan region.