Adaptive Sampling Mask for MRI Image Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional medical image acquisition techniques, such as MRI, suffer from slow acquisition speed, particularly for dynamic organs, leading to extended acquisition times and lower quality reconstructed images due to sub-optimal sampling masks.
Innovation Solution
A computer-implemented method and system that generates adaptive sampling masks using a model based on prior phase data and images, allowing for improved k-space data acquisition and image reconstruction, reducing acquisition time while enhancing image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional fixed sampling masks are used for image acquisition, then the acquisition process is simple and fast to implement, but the image quality deteriorates and pertinent information is missing
Solution Approach 1:
The patent applies dynamics by transitioning from fixed, static sampling masks to adaptive, dynamic sampling masks that change based on real-time image data. The system continuously adjusts sampling patterns according to the specific characteristics of each imaging scenario, making the sampling process adaptive rather than predetermined. This resolves the contradiction by allowing complex, optimized sampling patterns to be applied dynamically without requiring complex device architecture.
Solution Approach 2:
The patent implements feedback by using reconstructed images from preliminary scans to inform and optimize subsequent sampling mask generation. The system analyzes the initial image quality and structural information, then feeds this back into the sampling mask design process to create optimized masks that target specific regions of interest. This feedback loop enables high-quality imaging without requiring inherently complex sampling patterns from the start.
2Productivity
If conventional fixed sampling masks are used, then the acquisition setup is straightforward, but the acquisition time increases for dynamic organs
Solution Approach 1:
The patent applies preliminary action by performing a quick initial scan to generate preliminary images before the main imaging sequence. This preliminary action provides essential structural information that guides the subsequent optimized sampling process, allowing the system to skip unnecessary sampling steps and focus resources on critical regions. This preliminary step reduces overall acquisition time despite the added initial scan, because it prevents wasted time on redundant sampling in the main sequence.
Solution Approach 2:
The system dynamically adjusts sampling rates and patterns based on the specific organ being imaged and its motion characteristics. For dynamic organs like the heart, the sampling mask adapts to the organ's motion cycle, concentrating sampling efforts during critical phases while reducing sampling during less critical periods. This dynamic adaptation increases effective acquisition speed without sacrificing essential data collection.
3Manufacturing precision
If conventional fixed sampling masks are applied, then the sampling process is simple to implement, but the reconstructed image quality deteriorates
Solution Approach 1:
The patent implements self-service by enabling the sampling mask generation system to automatically optimize masks based on the specific imaging scenario without requiring manual intervention or complex pre-programming. The system uses the acquired image data itself to generate appropriate sampling masks, making the process self-adapting and self-optimizing. This reduces the need for complex external control systems while achieving high-quality reconstruction.
Solution Approach 2:
The patent applies parameter changes by dynamically adjusting sampling mask parameters such as sampling density, pattern orientation, and region of interest based on the analyzed image characteristics. Rather than using fixed complex patterns, the system modifies sampling parameters in real-time to match the specific anatomical features and diagnostic requirements, achieving high-quality reconstruction with adaptively simplified sampling strategies.
Data Source
AI summary
Methods and systems for acquiring a visualization of a target. For example, a computer-implemented method for acquiring a visualization of a target includes: generating a first sampling mask; acquiring first k-space data of the target at a first phase using the first sampling mask; generating a first image of the target based at least in part on the first k-space data; generating a second sampling mask using a model based on at least one selected from the first sampling mask, the first k-space data, and the first image; acquiring second k-space data of the target at a second phase using the second sampling mask; and generating a second image of the target based at least in part on the second k-space data.