Bone Segmentation via Iterative Axis Recalculation in 3D Extremity Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Cone-Beam Computed Tomography (CBCT) systems face challenges in accurately segmenting and measuring bone structures, particularly in extremities like hands, feet, and knees, due to complex spatial arrangements and varying bone shapes, which hinders precise morphometric characterization and diagnosis.
Innovation Solution
A method involving the acquisition of reconstructed tomographic volume images, construction of a primary axis for bones by sectioning and recalculating the axis through iterative processing, and displaying the recalculated axis for improved segmentation and measurement of bone structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional CBCT imaging is used for extremity imaging, then 3D volume images can be obtained, but accurate segmentation and feature identification for bone structures remain difficult due to complex spatial arrangements and varying bone shapes
Solution Approach 1:
The patent applies segmentation by dividing the bone structure analysis into distinct computational steps: initial axis estimation, orthogonal plane sectioning, and iterative axis recalculation. Each step processes a specific aspect of the bone geometry, breaking down the complex segmentation problem into manageable segments that can be solved sequentially to achieve accurate bone structure identification
Solution Approach 2:
The patent transitions from 2D cross-sectional views to 3D spatial reasoning by constructing and iteratively refining a primary axis that extends through the bone structure. This dimensional approach allows the system to capture the three-dimensional morphology and spatial orientation of bones, enabling accurate segmentation despite complex arrangements and varying shapes
2Measurement precision
If iterative axis construction and recalculation is performed, then bone segmentation accuracy is improved, but processing time and computational complexity increase
Solution Approach 1:
The patent performs preliminary action by first estimating an initial primary axis before iterative refinement. This initial estimation provides a starting point that is already reasonably accurate, reducing the number of iterations needed to converge on the final precise axis. The preliminary axis construction includes identifying key anatomical landmarks and establishing the general orientation, which speeds up subsequent refinement steps
Solution Approach 2:
The patent implements feedback through iterative recalculation of the primary axis based on sectioned bone images. Each iteration uses the results from the previous iteration to refine the axis position and orientation, continuously improving accuracy. The feedback loop compares the current axis estimate against the actual bone geometry and adjusts accordingly, ensuring high precision while converging efficiently
Data Source
AI summary
A method acquires a reconstructed tomographic volume image of anatomy of a patient and constructs a primary axis for a bone of interest in the imaged anatomy. The method forms a sectioned image of the bone in the volume image according to a first plane that is defined to extend along the bone and to extend through two or more articular surfaces. A primary axis for the bone is estimated, wherein the primary axis is midway between outer edges of the bone image that intersect the first sectioning plane. The bone is sectioned in the volume image by a second plane that is orthogonal to the first plane and that extends through the estimated primary axis. The method recalculates the position of the constructed primary axis according to the sectioning of the bone by the second plane. The recalculated constructed primary axis for the bone of interest is displayed.