Automated 3D Spinal Assessment via Model Registration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for assessing spinal deformity, such as the Cobb angle, are insufficient for accurately quantifying three-dimensional curvatures and rotations in scoliosis, and are prone to observer variability, necessitating the development of automated and reliable methods for evaluating anatomical features of the spine.
Innovation Solution
An automated method using a 3-D model with landmarks is applied to patient image data, allowing for precise measurements of vertebrae position, rotation, and deformation, including the calculation of Cobb angle, axial vertebral rotation, and local deformity, through affine and deformable registration techniques, enabling the analysis of spinal curvature and classification of scoliosis sub-types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the Cobb angle measurement method is used to assess spinal deformity, then the assessment can be performed using standard radiographs, but the measurement fails to accurately quantify three-dimensional curvatures and rotations
Solution Approach 1:
The patent transitions from two-dimensional Cobb angle measurements on radiographs to three-dimensional assessments using CT or MRI volumetric data. The system registers 3D spinal models to patient-specific volumetric image data, enabling measurement of axial vertebral rotation, lateral displacement, and sagittal/deformations in all three spatial dimensions, thereby resolving the limitation of the Cobb angle's inability to capture 3D deformity characteristics
Solution Approach 2:
The patent replaces manual measurement methods with an automated computer-based system that performs landmark identification, model registration, and deformity calculation automatically. The system uses algorithms to register pre-defined spinal models to patient-specific volumetric data and compute three-dimensional parameters, eliminating observer variability and reducing manual measurement complexity
2Reliability
If manual measurement methods are used for assessing spinal deformity, then the process can be performed with simple tools, but observer variability significantly affects measurement reliability
Solution Approach 1:
The system performs self-service by automatically identifying vertebral landmarks, registering spinal models to patient data, and calculating three-dimensional deformity parameters without requiring manual measurement. The automated algorithm consistently identifies anatomical features and computes measurements, eliminating observer variability while maintaining measurement reliability across different users and time points
Solution Approach 2:
The system incorporates feedback mechanisms where the registered spinal model is compared against the actual patient volumetric data, and measurement results are validated against established anatomical relationships. This feedback loop ensures measurement accuracy and reliability while maintaining automation
3Measurement precision
If three-dimensional image data is used for automated spinal assessment, then accurate measurement of axial rotation and lateral displacement is achieved, but the complexity of image processing increases
Solution Approach 1:
The system uses pre-defined spinal models with anatomically correct vertebral shapes and pre-identified landmarks as a starting point for assessment. These models are registered to patient-specific volumetric data before detailed measurements are performed, simplifying the processing complexity by establishing a known reference framework in advance
Solution Approach 2:
The patent segments the complex three-dimensional assessment task into distinct components: (1) extraction of subvolumes containing individual vertebrae from volumetric data, (2) registration of pre-defined spinal models to patient data, (3) identification of anatomical landmarks on each vertebra, and (4) calculation of specific deformity parameters. This segmentation reduces overall system complexity by breaking down the processing into manageable, modular steps
Data Source
AI summary
Systems, methods, computer programs, and circuits can perform automated spinal assessments using a 3-D model and a 3-D patient image data set of the spine which do not rely on vertebral symmetry.