Optical Scan Spinal Shape Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current diagnostic methods for scoliosis, such as radiography, expose patients to ionizing radiation and provide only two-dimensional projections of three-dimensional spinal structures, lacking accuracy for non-invasive diagnosis.
Innovation Solution
A method and system that utilize three-dimensional (3D) surface scans and trained neural network algorithms to predict spinal shape, enabling accurate estimation of skeletal deformities like scoliosis without ionizing radiation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If radiography is used for scoliosis diagnosis, then accurate spinal structure information is obtained, but patients are exposed to ionizing radiation and only 2D projections are provided
Solution Approach 1:
The patent replaces the radiographic imaging system (which uses ionizing radiation) with an optical scanning system that uses light to capture 3D surface geometry. The optical scanner projects patterns and captures images to reconstruct the spinal surface, substituting mechanical/radiation-based imaging with optical field-based imaging that is safe for repeated use
Solution Approach 2:
The patent transforms the 2D projection data from radiography into 3D surface geometry data through optical scanning. By capturing the spinal surface from multiple angles and reconstructing the three-dimensional shape, the system provides dimensional information that was lost in 2D projections, enabling accurate curvature measurement without radiation
2Loss of information
If radiography is used for scoliosis diagnosis, then spinal structure is visualized, but only two-dimensional projection of three-dimensional structure is provided
Solution Approach 1:
The patent explicitly addresses the dimensionality loss by using optical scanning to capture 3D surface geometry. The system reconstructs the spinal surface as a three-dimensional model from multiple 2D images, preserving all spatial information including depth, allowing accurate measurement of curvatures and rotations in three dimensions rather than flattened 2D projections
3Object-affected harmful factors
If optical scan is used to estimate spinal shape, then no ionizing radiation exposure occurs, but accurate spinal shape prediction requires complex neural network training
Solution Approach 1:
The patent applies preliminary action by training the neural network model in advance using large datasets of paired optical scans and radiographic images. The model learns the mapping from surface geometry to spinal shape during an offline training phase, so that during actual diagnosis, the pre-trained model can quickly predict spinal shape from new optical scans without requiring complex real-time computation or additional radiation exposure
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method comprising: generating a parametrized three-dimensional (3D) body surface model on a training set comprising a plurality of 3D scans of subjects, wherein at least some of said 3D scans are of subjects having a skeletal deformity; receiving one or more target 3D scans of a target subject; optimizing said body surface model with respect to said one or more target 3D scans to calculate a target body surface model of said target subject; training a skeletal estimation model on a training set comprising: (i) body surface models of a plurality of subjects, and (ii) skeletal landmarks sets of said plurality of subjects; and applying said trained skeletal estimation model to said calculated target body surface model of said target subject, to estimate a skeletal shape of said target subject.