Patient Scan Monitoring With Perspective-Corrected Motion Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing imaging systems struggle to accurately track and monitor a patient's region of interest during medical scans due to perspective changes and occlusions, especially when the patient is moved through a medical imaging device, making it difficult to distinguish between apparent motion and true patient motion, which affects image quality and safety.
Innovation Solution
An imaging system with cameras mounted orthogonally to the patient support displacement, performing image post-processing to correct distortions and perspective, allowing for patient motion analysis in the pixel column direction, and identifying regions of interest for precise motion detection and tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If cameras are mounted on room walls or ceiling to capture patient views, then the scanning system can be accessed, but the view of the patient is partially occluded and incomplete during scans
Solution Approach 1:
A transparent or translucent monitor screen is positioned within the bore of the scanning system to serve as an intermediary viewing surface. The screen displays real-time camera images of the patient, allowing operators to monitor patient status without requiring direct line-of-sight cameras mounted on walls. This intermediary display system eliminates occlusion issues while maintaining continuous monitoring capability throughout the scanning procedure.
Solution Approach 2:
The monitoring system transitions from traditional 2D wall-mounted camera views to a 3D spatial arrangement where cameras are positioned at multiple locations (front, rear, and side views) and their images are synthesized on a transparent monitor screen within the bore. This dimensional reconfiguration allows complete patient coverage without occlusion by positioning viewing surfaces in the scanning path rather than around it.
2Stability of the object's composition
If cameras are mounted on the patient support to provide stable view, then monitoring stability improves, but the view is obstructed by patient body parts and medical devices
Solution Approach 1:
Instead of using a single camera position that would be obstructed by patient body parts, the system segments the monitoring function into multiple cameras positioned at different locations (front, rear, and side views of the patient support). Each camera captures a specific region, and the images are synthesized together on the transparent monitor screen, providing complete patient coverage without obstruction from any single viewpoint.
Solution Approach 2:
The transparent monitor screen acts as an intermediary that combines images from multiple cameras positioned around the patient support. This allows the system to achieve stable monitoring by having cameras at optimal positions while using the screen as a virtual mounting point that is not obstructed by patient body parts or medical devices.
3Object-affected harmful factors
If cameras are mounted on the scanning system gantry facing the patient, then the view is not occluded, but separating apparent motion from true patient motion becomes difficult
Solution Approach 1:
The monitoring system segments the view into multiple perspectives by placing cameras at different locations (front, rear, and side views) rather than using a single gantry-mounted camera. This segmentation allows the system to distinguish between motion components by analyzing how different regions of the patient move relative to each other across multiple camera views, improving motion detection precision while maintaining complete visibility.
Solution Approach 2:
The system adds a temporal dimension to the spatial arrangement by capturing images from multiple camera positions simultaneously and synthesizing them on a transparent monitor screen. This multi-dimensional approach (multiple spatial views plus time) enables the system to separate apparent motion from true patient motion through image processing algorithms that analyze motion patterns across all views.
Data Source
AI summary
An imaging system is for capturing images of a patient during a medical scan. A set of one or more cameras captures images of the patient such that, after applying distortion and perspective correction steps, pixel rows correspond to locations along an axis parallel to the direction of patient support movement and pixel columns correspond to locations along an axis perpendicular to the patient support. Patient motion of interest can then be identified based on image movements in the pixel column direction between sequential captured images. A measure of patient motion can then be derived.