3D Dental Impression Fusion for CBCT Metal Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Metal artifacts in Cone Beam Computed Tomography (CBCT) images, such as beam hardening, scatter, and photon starvation artifacts, distort dental images, leading to inaccurate diagnosis and treatment planning, and existing methods like iterative reconstruction and deep learning require extensive datasets or increase X-ray dose.
Innovation Solution
A method and system for metal artifact reduction in dental scan images via post-processing using surface data from dental digital impressions, involving alignment, segmentation, merging, and thickening of 3D digital impressions to form a shell around distorted boundaries, effectively eliminating streaks and dark regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If iterative reconstruction or dual energy-based methods are used to reduce metal artifacts, then artifact reduction is achieved, but X-ray dose to the patient increases
Solution Approach 1:
The patent applies preliminary action by performing metal artifact reduction through post-processing of the CBCT images using deep learning algorithms. Instead of modifying the imaging process itself (which would increase X-ray dose), the method processes the already-acquired images to remove artifacts, thereby avoiding additional radiation exposure while still achieving artifact reduction
Solution Approach 2:
The patent extracts and removes the harmful metal artifact components from the CBCT images using deep learning-based segmentation and reconstruction algorithms. The method identifies and separates artifact regions from genuine anatomical structures, eliminating the harmful effects without requiring additional X-ray exposure
2Object-affected harmful factors
If deep learning-based methods are used to reduce metal artifacts, then artifact reduction is achieved, but enormous dataset of patient images is required for training
Solution Approach 1:
The patent performs preliminary segmentation and identification of metal structures and artifact regions before final image reconstruction. This preliminary processing step allows the deep learning model to focus on specific artifact patterns rather than requiring training on complete patient datasets, reducing the training data requirement while maintaining effective artifact reduction
Solution Approach 2:
The patent segments the CBCT images into different regions (metal structures, artifact regions, and genuine anatomical structures) using deep learning algorithms. This segmentation approach allows the system to learn from smaller, targeted datasets focused on artifact patterns rather than requiring enormous complete patient datasets
3Reliability
If metal artifacts are present in CBCT images, then imaging of metallic structures is achieved, but image readability and alignment accuracy deteriorate
Solution Approach 1:
The patent converts the harmful effect of metal artifacts into a beneficial process by using the deep learning model to identify and characterize artifact patterns. The same metallic structures that cause artifacts provide training data and reference points for the algorithm to learn artifact patterns, which are then removed to improve both image readability and alignment accuracy
Data Source
AI summary
A method includes acquiring a first volumetric image data set representing dentition of a patient, the first volumetric image data set including a modeled structure having an artifact distorting a boundary thereof, aligning a 3D digital impression of dentition of the patient with at least a portion of the first volumetric image data set, segmenting individual structures in the first volumetric image data set, merging the segmented individual structures to form a unitary volumetric model, thickening the 3D digital impression to form a shell bounding the at least a portion of the first volumetric image data set and supplementing the boundary of the at least one modeled structure, and, combining the 3D digital impression and the unitary volumetric model into a second volumetric image data set including the unitary volumetric model having at least a portion thereof bounded by the 3D digital impression.