MRI Image Processing Apparatus Dual-Filter Noise Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Magnetic resonance imaging (MRI) systems face challenges in accurately calculating flow velocity and improving image quality due to the limitations of existing noise removal methods, which either fail to correctly eliminate noise or introduce errors when using the same filters for magnitude and phase images.
Innovation Solution
The implementation of different filters for magnitude and phase images, where the magnitude image generator applies a stronger low-pass filter to remove direct current components and the phase image generator uses a weaker filter to preserve higher frequency components, allowing for accurate flow velocity calculation and improved image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the same filter is applied to both magnitude and phase images, then the processing is simple, but the flow velocity calculation becomes inaccurate and image quality deteriorates
Solution Approach 1:
The patent divides the filtering process into two separate paths: one for magnitude images and one for phase images. Each path uses a filter optimized for its specific requirements, with the magnitude image path using a filter that removes DC components and the phase image path using a filter that preserves higher frequency components necessary for flow velocity calculation.
Solution Approach 2:
The patent applies different filtering characteristics to different image types based on their specific requirements. Magnitude images receive stronger low-pass filtering to remove noise and DC components, while phase images receive weaker filtering to preserve the higher frequency components that contain flow velocity information.
2Reliability
If a strong low-pass filter is applied to remove noise, then the signal-to-noise ratio improves, but the flow velocity information is lost
Solution Approach 1:
The patent separates the noise removal process into two distinct filtering paths. The magnitude image path applies a strong low-pass filter to achieve high signal-to-noise ratio, while the phase image path applies a weaker filter that preserves the frequency components containing flow velocity information.
Solution Approach 2:
The patent changes the filter parameters (cut-off frequency, filter strength) based on the specific requirements of each image type. For magnitude images, the filter parameters are set to maximize noise removal, while for phase images, the parameters are adjusted to preserve flow velocity information.
3Manufacturing precision
If DC components are removed from magnitude images, then image quality improves, but the processing becomes more complex
Solution Approach 1:
The patent implements a dedicated filtering path for magnitude images that specifically targets DC component removal. This segmented approach allows the system to apply DC removal only where needed for magnitude images without affecting the phase image processing pipeline.
Solution Approach 2:
The patent adjusts the filter parameters in the magnitude image path to specifically target and remove DC components while preserving the useful signal. This parameter optimization improves image quality by eliminating unwanted DC offsets without unnecessarily complicating the overall processing system.
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 enables the correct calculation of flow velocity and enhances the image quality of MRI images by effectively removing noise specific to each image type, improving the signal-to-noise ratio and accuracy of flow velocity data.
Implementation Method 1
A magnetic resonance imaging apparatus (MRI apparatus) is an apparatus that images internal information of a subject by using a nuclear magnetic resonance phenomenon. An MRI apparatus acquires, with a coil, data referred to as k-space data by sampling nuclear magnetic resonance signals (MR signals) from certain atomic nuclei
Implementation Method 2
acquires an MR image by applying the Fourier transform to the k-space data
Data Source
AI summary
An image processing apparatus according to an embodiment includes conversion circuitry, magnitude image generating circuitry and phase image generating circuitry. The conversion circuitry is configured to convert time-series k-space data into first time-series x-space data, the x-space representing a spatial position. The magnitude image generating circuitry is configured to generate a magnitude image from second time-series x-space data, the second time-series x-space data being acquired by applying a first filter to the first time-series x-space data. The phase image generating circuitry is configured to generate a phase image from third time-series x-space data, the third time-series x-space data being acquired by applying, to the first time-series x-space data, a second filter that is different from the first filter.