Statistical Shape Modeling for Pre-Morbid Anatomy Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing surgical joint repair procedures face challenges in accurately determining pre-morbid characteristics of patient anatomy due to damage or disease progression, which affects proper prosthetic selection and positioning, as current imaging techniques suffer from under- and over-segmentation, leading to loss of usable data.
Innovation Solution
Utilizing statistical shape modeling (SSM) to determine a pre-morbid shape of an anatomical object by aligning segmented patient anatomy to an initial shape model through iterative processes like ICP and elastic registration, adjusting size, shape, and location to generate a pre-morbid approximation despite under- and over-segmentation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If imaging techniques are used to capture patient anatomy, then visual data is obtained, but under- and over-segmentation causes loss of usable data
Solution Approach 1:
A statistical shape model serves as an intermediary between the segmented anatomical data and the desired pre-morbid characterization. The model incorporates prior knowledge of normal anatomical variations and uses the available (albeit imperfect) segmented data to infer the complete pre-morbid anatomy, effectively mediating the information loss caused by segmentation errors
Solution Approach 2:
The system changes parameters by adjusting the statistical shape model to fit the segmented data while maintaining anatomical plausibility. By varying model parameters within statistically valid ranges, the system recovers lost anatomical information and compensates for segmentation inaccuracies
2Reliability
If damage or disease progression is present, then current anatomy can be imaged, but pre-morbid characteristics are lost
Solution Approach 1:
The statistical shape model is pre-populated with knowledge of normal anatomical characteristics and variations before the imaging process. This preliminary preparation allows the model to infer and reconstruct pre-morbid characteristics even when only post-damage anatomy is visible in the imaging data
Solution Approach 2:
The system creates a copy or representation of the pre-morbid anatomy through the statistical shape model, rather than directly observing it. This modeled copy preserves the essential pre-morbid characteristics by leveraging statistical relationships from population data, enabling surgical planning as if the pre-morbid state were directly visible
3Measurement precision
If rigid transformation and non-rigid registration are performed, then anatomical alignment is achieved, but computational complexity increases
Solution Approach 1:
A rigid transformation is performed as a preliminary step before non-rigid registration, establishing a coarse alignment that reduces the search space for subsequent detailed registration. This preliminary action simplifies the overall computational complexity by breaking down the complex registration task into manageable stages
Solution Approach 2:
The registration process is segmented into distinct stages: rigid transformation followed by non-rigid registration. This segmentation allows each stage to focus on specific aspects of alignment, reducing computational complexity compared to attempting a single comprehensive registration step
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
Techniques are described for determining a pre-morbid shape of an anatomical object. Processing circuitry may determine an aligned shape based on image data with under- or over-segmentation. The processing circuitry may utilize the aligned shape and a shape model such as a mean shape model of the anatomical object to register the aligned shape to the shape model and generate information indicative of the pre-morbid shape of the anatomical object.