Aortic Wall Stress Prediction From Neural Network 3D Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for simulating wall stress of body parts, such as the aorta, are laborious and computationally exhaustive, requiring manual generation of three-dimensional models and simulation data, which is time-consuming.
Innovation Solution
A system utilizing neural networks to automatically detect boundaries and predict wall stress by generating three-dimensional models from cross-sectional images, reducing computational resources and time through machine learning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual generation of three-dimensional models and simulation data is used, then measurement precision and manufacturing precision are improved, but loss of time and device complexity increase significantly
Solution Approach 1:
The patent creates a digital copy of the manual simulation process through a neural network model. The neural network learns from manually generated training data (ground truth) and reproduces wall stress predictions without requiring manual model generation for each new case, thus maintaining accuracy while eliminating repetitive manual work
Solution Approach 2:
The patent replaces the manual mechanical process of generating three-dimensional models and running simulations with an automated neural network system. The neural network substitutes for the complex computational mechanics process, providing rapid predictions without requiring manual intervention in the simulation workflow
2Manufacturing precision
If manual generation of three-dimensional models is used, then manufacturing precision is improved, but productivity deteriorates due to laborious process
Solution Approach 1:
The patent performs preliminary action by pre-training the neural network on manually generated data during an initial phase. Once trained, the neural network can rapidly predict wall stress for new cases without requiring manual model generation for each prediction, thus maintaining manufacturing precision while dramatically improving productivity
Solution Approach 2:
The neural network creates a computational copy of the precise manual modeling and simulation process. This copy reproduces the accuracy of manual methods but executes much faster, eliminating the bottleneck between model generation and stress computation
3Reliability
If computational simulation is used, then reliability is improved, but use of energy and device complexity increase
Solution Approach 1:
The patent creates a lightweight neural network model that copies the essential predictive capability of complex computational simulations. This neural network copy requires significantly less computational energy to run while maintaining the reliability of wall stress predictions, as it has already learned the complex relationships during training
Data Source
AI summary
Disclosed herein are related to a system, a method, and a non-transitory computer readable medium for simulating, predicting, or estimating, based on machine learning neural networks, wall stress of a body part. In one approach, a first neural network automatically detects features in multiple images of a body part. For example, the first neural network may detect, for each image, a lumen and a wall of an aorta. According to the detected features, a second neural network may simulate, estimate, or predict wall stress of the body part in response to pressure applied to the body part. For example, a model generator can generate a three-dimensional model of the body part according to the detected features in the multiple images, and the second neural network can simulate, estimate, or predict wall stress of the body part according to the three-dimensional model.