Adaptive CT Metal Segmentation for Artifact-Free Anatomy Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Visual artifacts caused by metal objects in computed tomography (CT) images degrade the quality of reconstructed patient anatomy, hindering accurate detection of target volumes and critical structures during radiation therapy.
Innovation Solution
A computer-implemented method employing automated metal extraction, inpainting, blending, and metal object restoration techniques to reduce or eliminate visual artifacts in reconstructed volumes, using adaptive autosegmentation and harmonic functions to ensure smooth transitions and accurate image reconstruction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If metal objects are present in the scanned region, then radiation therapy treatment can be delivered, but visual artifacts are generated that degrade image quality and hinder accurate detection of target volumes
Solution Approach 1:
The patent extracts metal objects from the reconstructed volume by generating a metal mask that identifies metal regions, then removes these regions to create a metal-free reconstructed volume. This extraction process eliminates the source of visual artifacts while preserving the underlying anatomical structures through inpainting techniques.
Solution Approach 2:
The patent introduces a metal mask as an intermediary element that separates metal objects from the reconstructed volume. The mask serves as a mediator to identify and isolate metal regions, enabling subsequent removal and inpainting operations to eliminate artifacts while maintaining image quality.
2Measurement precision
If conventional thresholding methods are used for metal object extraction, then the process is simple, but accuracy is insufficient due to inability to differentiate metal from high-density tissue
Solution Approach 1:
The patent applies local quality by using a gradient-based algorithm that analyzes local intensity variations around potential metal objects. The algorithm calculates gradient magnitudes and directions to distinguish metal edges from tissue edges, applying different extraction criteria to different regions based on their local characteristics.
Solution Approach 2:
The patent changes multiple parameters including gradient threshold values, search radius, and intensity difference thresholds to optimize metal object detection. These parameter adjustments enable the algorithm to adapt to different metal types, sizes, and densities while maintaining high extraction accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented method of segmenting a reconstructed volume of a region of patient anatomy includes: determining 703 an anatomical region associated with the reconstructed volume; detecting 705 one or more metal objects disposed in an initial 3D metal object mask associated with the reconstructed volume; for each of the one or more metal objects disposed in the initial 3D metal object mask, determining a volume associated with the metal object; determining 711 a value for at least one segmentation parameter based on the anatomical region and on the volume associated with the one or more metal objects; and generating 713 a final 3D metal object mask associated with the reconstructed digital volume using the value for the segmentation parameter.