ML-Based Imaging Geometry Adjustment for Interventional X-Ray
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current interventional X-ray imaging systems lack awareness of patient position on the examination table, requiring manual and time-consuming adjustments of imaging geometry, leading to increased procedure duration and X-ray dose.
Innovation Solution
A system utilizing a pre-trained machine learning component to compute and automate imaging geometry changes based on live imagery, predicting the required adjustments to switch between regions of interest and avoiding collisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual adjustment of imaging geometry is used, then the system is simple to operate, but procedure time increases and productivity decreases
Solution Approach 1:
The system automatically detects the patient's body part position and autonomously adjusts imaging geometry parameters without requiring manual operator intervention. The imaging system serves itself by implementing self-positioning and self-adjustment capabilities through automated control algorithms that respond to detected anatomical landmarks.
Solution Approach 2:
The patent replaces manual mechanical adjustment operations with an automated computer-controlled system. The mechanical movement of the imaging device and examination table is substituted by an automated control system that calculates and executes positioning based on detected anatomical features, eliminating the need for manual mechanical manipulation.
2Device complexity
If manual adjustment of imaging geometry is used, then device complexity is low, but automation extent is insufficient
Solution Approach 1:
The imaging system integrates multiple functions into a single automated platform: anatomical detection, position recognition, geometry calculation, and automated positioning control. The system universally handles different body parts and imaging scenarios through a unified automated control architecture, replacing multiple manual operations with one integrated intelligent system.
Solution Approach 2:
The patent introduces an automated control system as an intermediary between the operator and the imaging device. This intermediary layer processes detected anatomical information and automatically generates positioning commands, serving as a mediator that translates anatomical detection results into precise imaging geometry adjustments without requiring direct manual intervention.
3Reliability
If X-ray imaging continues during table motion to monitor FOV, then imaging coverage is maintained, but X-ray dose increases
Solution Approach 1:
The system performs preliminary detection of anatomical landmarks and pre-calculates the required imaging geometry adjustments before initiating table or C-arm motion. By determining the optimal imaging parameters in advance based on detected body part positions, the system eliminates the need for continuous X-ray imaging during motion, thereby reducing unnecessary radiation exposure while maintaining imaging coverage.
Solution Approach 2:
The patent implements rapid automated positioning that quickly transitions the imaging system to the target position without requiring prolonged intermediate imaging monitoring. The accelerated motion and precise pre-planned trajectory allow the system to skip through the positioning process efficiently, minimizing the time during which X-ray imaging would be required to monitor field of view.
4Manufacturing precision
If slow table motion is used for precise positioning, then positioning precision is improved, but procedure time increases
Solution Approach 1:
The system pre-calculates the exact positioning requirements and optimal motion trajectory based on detected anatomical landmarks before initiating table motion. By determining the precise target position and path in advance, the system can execute faster motion with confidence in achieving accurate positioning, eliminating the need for slow incremental adjustments.
Solution Approach 2:
The patent implements a feedback mechanism where the system continuously monitors the actual position during motion and compares it with the pre-calculated target position. This real-time feedback allows the system to make minor corrections if needed while maintaining overall faster motion speeds, achieving both precision and efficiency through closed-loop control.
Data Source
AI summary
System (SYS) and delated methods for supporting an imaging operation of an imaging apparatus (IA) capable of assuming different imaging geometries. The system comprises an input interface (IN) for receiving a current image acquired by the imaging apparatus (IA) of a current region of interest (ROI_1) at a current imaging geometry (p). A pre-trained machine learning component (MLC) computes output data that represents an imaging geometry change (Δp) for a next imaging geometry (p′) in relation to a next region of interest (ROI_2). An output interface (OUT) outputs a specification of the imaging geometry change.