ML-Adjusted C-Arm CBCT Trajectories for Metal Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current spinal fusion surgery techniques face challenges due to high rates of misplaced pedicle screws, with cortical breach occurring in up to 31% and 72% of cases for freehand and fluoroscopy-guided techniques respectively, leading to nerve damage and suboptimal image quality from C-arm cone-beam computed tomography (CBCT) devices.
Innovation Solution
A surgical system utilizing a machine learning model to adjust CBCT device trajectories for artifact avoidance by processing X-ray images to predict the quality of next possible images and updating the six-degree of freedom pose to minimize image corruption from metal artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If C-arm CBCT device uses traditional fixed trajectory for imaging, then device operation is simple, but image quality deteriorates due to metal artifacts and noise
Solution Approach 1:
The patent implements dynamic trajectory adjustment by modifying the C-arm CBCT device's scanning path in real-time based on detected metal artifact patterns. The trajectory transitions from a fixed predetermined path to a dynamically adapted path that avoids regions causing severe artifacts, thereby improving image quality without requiring complete redesign of the imaging system.
Solution Approach 2:
The system incorporates feedback mechanisms where image quality metrics are continuously evaluated during scanning, and trajectory adjustments are made based on this feedback. The machine learning model analyzes incoming images for metal artifact presence and feeds this information back to modify subsequent trajectory points, creating a closed-loop control system that optimizes image quality iteratively.
2Loss of information
If C-arm CBCT device captures images at multiple fixed positions, then image coverage is comprehensive, but processing time increases due to artifact correction requirements
Solution Approach 1:
The system performs preliminary actions by pre-planning an optimized trajectory that anticipates and avoids metal artifact-prone regions before imaging begins. The machine learning model pre-processes anatomical information to identify potential artifact sources and adjusts the trajectory in advance, preventing artifact formation rather than correcting it later, thus reducing processing time while maintaining information completeness.
Solution Approach 2:
The patent changes imaging parameters dynamically along the trajectory, including angular positions, scanning speeds, and exposure settings, to optimize image acquisition. By varying these parameters adaptively rather than using fixed settings, the system maintains comprehensive coverage while reducing the number of images requiring heavy artifact correction, thereby decreasing processing time.
3Reliability
If surgeon uses freehand or fluoroscopy-guided technique for screw placement, then operation procedure is simple, but cortical breach rate increases to 31-72%
Solution Approach 1:
The patent introduces an intermediary system consisting of the machine learning model and optimized trajectory planning that mediates between the surgeon's intent and the actual screw placement. This intermediary provides real-time guidance by identifying optimal imaging angles and trajectories that clearly visualize anatomical landmarks and screw paths, thereby improving placement accuracy without requiring the surgeon to perform complex manual adjustments.
Solution Approach 2:
The system replaces purely mechanical freehand techniques with an intelligent guidance system that uses machine learning algorithms to determine optimal imaging and drilling trajectories. This substitution transforms the manual trial-and-error approach into a computationally-guided process, significantly reducing cortical breach rates while maintaining ease of operation through automated trajectory recommendation.
4Measurement precision
If C-arm CBCT device follows standard circular trajectory, then device operation is straightforward, but metal artifacts corrupt image data significantly
Solution Approach 1:
The patent applies parameter changes by modifying trajectory parameters (angular positions, radial distances, scanning velocities) from the standard circular pattern to an optimized non-circular path. The machine learning model determines specific parameter adjustments that minimize metal artifact formation while maintaining adequate image coverage, thereby improving data quality without requiring complex manual programming.
Solution Approach 2:
The system implements self-service by enabling the C-arm CBCT device to automatically adjust its own trajectory based on real-time image analysis and machine learning predictions. The device autonomously identifies optimal imaging paths and executes trajectory modifications without external intervention, reducing the programming complexity burden on operators while significantly improving image data quality through artifact-minimized pathways.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system effectively reduces metal artifacts and noise in CBCT images, improving the assessment of cortical breach and reducing the risk of nerve damage by autonomously adjusting CBCT device trajectories during spinal fusion surgery.
Implementation Method 1
processing the X-ray image, with a machine learning model, to determine a predicted quality of next possible X-ray images provided by the C-arm CBCT device
Implementation Method 2
receiving an X-ray image captured by a C-arm cone-beam computed tomography (CBCT) device at a particular position defined by a six-degree of freedom pose relative to an anatomy
Implementation Method 3
C-arm cone-beam computed tomography (CBCT) device
Data Source
AI summary
A device may receive an X-ray image captured by a C-arm CBCT device at a particular position defined by a six-degree of freedom pose relative to an anatomy, and may process the X-ray image, with a machine learning model, to determine a predicted quality of next possible X-ray images provided by the C-arm CBCT device. The device may utilize the machine learning model, to identify a particular X-ray image, of the next possible X-ray images, with a greatest predicted quality and to update the six-degree of freedom pose based on the particular X-ray image. The device may provide, to the C-arm CBCT device, data that identifies the updated six-degree of freedom pose to cause the C-arm CBCT device to adjust to a new position based on the updated six-degree of freedom pose.