Voxel-Wise Magnetic Susceptibility Mapping via Predictive Inverse Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current non-invasive characterization techniques, such as MRI, are time-consuming, expensive, and require long scan times, often involving large external magnetic fields that can be confining for patients, and determining model parameters using simulations is also challenging and resource-intensive.
Innovation Solution
A system that applies an external magnetic field and radio-frequency waves to a sample, measures the response, and uses an inverse model and predictive model to determine magnetic susceptibilities on a voxel-by-voxel basis, reducing the need for iterative measurements and improving characterization efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional MRI techniques are used to achieve high-spatial resolution, then measurement precision is improved, but scan time increases significantly
Solution Approach 1:
The patent applies preliminary action by using a predictive model to estimate model parameters before performing measurements. The predictive model uses training data and a forward model to predict what measurements should look like for given parameters, allowing the system to identify the minimum set of measurements needed to achieve desired precision without performing all possible scans.
Solution Approach 2:
The patent uses a predictive model that creates a virtual copy or simulation of the measurement process. By training on simulated or previous measurement data, the model can predict outcomes without physically performing each measurement, thereby reducing the actual scan time while maintaining precision through the simulated characterization.
2Measurement precision
If iterative measurements are performed to determine model parameters, then measurement precision is improved, but productivity decreases
Solution Approach 1:
The system performs preliminary characterization by using a predictive model to estimate model parameters before iterative measurements begin. The predictive model provides initial parameter estimates that can be used to guide subsequent measurements, reducing the number of iterations needed and thereby improving productivity while maintaining precision.
Solution Approach 2:
The patent implements feedback by using the predictive model to continuously refine parameter estimates. The model takes measured data and forward model predictions as input, compares them, and updates parameter estimates accordingly. This feedback loop allows the system to converge on accurate parameters more efficiently than traditional iterative methods, improving both precision and productivity.
3Measurement precision
If large external magnetic fields are used for MRI, then measurement precision is improved, but device complexity and patient comfort worsen
Solution Approach 1:
The patent uses a predictive model that creates a computational copy of the physical measurement process. Instead of relying solely on complex large-scale magnetic fields to achieve precision, the system uses the predictive model to simulate and characterize samples, thereby reducing the physical complexity of the magnetic field generation while maintaining measurement precision.
Solution Approach 2:
The patent replaces the mechanical/physical magnetic field system with a computational model. The predictive model uses algorithms and training data to characterize samples without requiring the same level of physical magnetic field strength, thereby reducing device complexity while maintaining or improving measurement precision through computational rather than purely physical means.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reduces MR scan times, increases throughput, improves user experience, and enhances the accuracy of physical parameter determination, allowing for faster and more cost-effective characterization of samples.
Implementation Method 1
magnetic properties can be studied using magnetic resonance or MR (which is often referred to as 'nuclear magnetic resonance' or NMR), a physical phenomenon in which nuclei in a magnetic field absorb and re-emit electromagnetic radiation
Implementation Method 2
These nuclear spins may precess or rotate around the direction of the external magnetic field at an angular frequency (which is sometimes referred to as the 'Larmor frequency')
Data Source
AI summary
A system may measure a response associated with a sample to an excitation. The system may compute, using the measured response and the excitation as inputs to an inverse model or a predetermined predictive model, model parameters on a voxel-by-voxel basis in a forward model with multiple voxels that represent the sample. The predetermined predictive model was trained using training data for different excitation strengths, different measurement conditions, or both. The forward model may simulate response physics occurring within the sample to a given excitation, and the model parameters may include magnetic susceptibilities of the multiple voxels. Moreover, the system may determine an accuracy of the model parameters by comparing at least the measured response and a calculated predicted value of the response using the forward model, the model parameters and the excitation. When the accuracy exceeds a predefined value, the system may provide the model parameters as an output.