ML-Based Device Deployment Simulation Using Signed Distance Fields
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for simulating the deployment of expandable devices in surgical planning are complex, time-consuming, and often oversimplify the device and anatomy, leading to inaccuracies and risks in surgical interventions.
Innovation Solution
A computer-implemented method using a machine learning system that calculates signed distance fields and geometrical parameters from medical images to generate accurate, fast, and reliable simulations of expandable device deployment within hollow structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex simulation techniques are used to simulate device deployment, then accuracy is improved, but computation time increases and real-time information cannot be provided
Solution Approach 1:
The patent pre-calculates signed distance fields and geometrical parameters from medical images before the actual surgical planning. This preliminary processing creates a simplified geometric model that can be quickly queried during surgical planning, eliminating the need for time-consuming simulations at the moment of decision-making.
Solution Approach 2:
The patent creates a simplified geometric copy of the patient-specific anatomy using signed distance fields and extracted geometrical parameters. This copy retains the essential geometric features needed for accurate device deployment simulation while being computationally efficient to process, replacing the need for complex full-physics simulations.
2Productivity
If over-simplifications of device and anatomy are applied, then computation time is reduced, but manufacturing precision and reliability deteriorate
Solution Approach 1:
The patent extracts specific geometrical parameters (curvature, diameter, length, cross-sectional area) from the patient's anatomy and uses these parameters to accurately represent the complex geometry in the simulation. This parameter-based approach maintains precision while enabling fast computation, avoiding both over-simplification and excessive complexity.
3Adaptability or versatility
If detailed geometric models are used for each patient, then personalized medicine standards are met, but data processing complexity and training time increase
Solution Approach 1:
The patent extracts only the essential geometrical parameters from detailed patient-specific medical images, separating the critical geometric features from the full complexity of the original images. This extraction process creates simplified yet patient-specific geometric models that reduce data processing complexity while maintaining the necessary level of detail for personalized surgical planning.
Data Source
AI summary
A method for training a machine learning system, including: based on at least one image dataset representing at least one portion of the hollow structure, calculating a signed distance field of each portion and calculating at least one geometrical parameter of each portion; generating a deployed in-use representation of the device by computing contact forces between the device and each portion based on the signed distance field and by applying these contact forces to a geometrical representation of the device; and training the machine learning system with each calculated geometrical parameter as an input and the corresponding deployed in-use representation as an associated target output, the obtained trained machine learning system being configured to receive as input at least one geometrical parameter and provide as output the deployed in-use representation of the device.