CT Image Noise Removal via Frequency Threshold Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current CT-image processing methods for X-ray CT imaging of animals, such as mice or rats, struggle to accurately capture heart and lung images due to noise from heartbeat and breathing motions, making it difficult to observe internal organs clearly, especially during intermediate phases between inflation and deflation periods.
Innovation Solution
A CT-image processing apparatus that uses threshold specification units to define lower limits for heartbeat and breathing frequencies, employing band-pass and high-pass filters to extract accurate waveforms, and reconfigures CT images at specific phases to remove noise and enhance image clarity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If projection data is captured during animal CT imaging, then internal organ images can be obtained, but noise from heartbeat and breathing motions blurs the images and reduces observation accuracy
Solution Approach 1:
The patent segments the motion signals by separating heartbeat and breathing beat frequencies through Fourier transform analysis. By identifying distinct frequency components, the method divides the composite motion signal into separable periodic components, allowing independent extraction and removal of each motion type's noise from the projection data.
Solution Approach 2:
The patent changes the parameter of frequency domain analysis by applying Fourier transform to convert time-domain motion signals into frequency-domain representations. This parameter transformation enables the identification and isolation of specific frequency components corresponding to heartbeat and breathing beats, facilitating selective noise removal while preserving image quality.
2Measurement precision
If dedicated apparatus such as ECG or video-camera is used for synchronization, then heartbeat or breathing beat can be obtained, but cost and operational burden increase
Solution Approach 1:
The patent makes the X-ray CT apparatus itself multi-functional by enabling it to extract both imaging data and synchronization signals (heartbeat and breathing beats) from the same projection data. This eliminates the need for separate dedicated apparatus like ECG machines or video cameras, as the CT system performs both imaging and physiological signal acquisition functions.
Solution Approach 2:
The patent implements self-service by having the CT imaging system automatically extract synchronization signals from its own projection data without requiring external dedicated equipment. The system uses its captured projection data to perform Fourier transform analysis and automatically identify heartbeat and breathing beat frequencies, making the apparatus self-sufficient for synchronization purposes.
3Measurement precision
If ROI tracking calculation is performed for synchronization, then synchronization signal can be obtained, but special setting and calculation complexity are required
Solution Approach 1:
The patent extracts synchronization signals directly from the existing projection data without requiring separate ROI tracking calculations. By applying Fourier transform to the projection data itself, the method extracts heartbeat and breathing beat frequency information directly, eliminating the need for additional ROI definition and tracking calculation steps.
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 effectively removes noise from heartbeat or breathing motions, allowing for the clear extraction of accurate periodic motions, thereby obtaining precise CT images at desired phases, including intermediate periods of lung or heart activity.
Implementation Method 1
a breathing beat extraction unit configured to extract a breathing beat waveform using a band-pass filter defined by the lower limit of the breathing beat frequency and the lower limit of the heartbeat frequency
Implementation Method 2
a heartbeat extraction unit configured to extract a heartbeat waveform using a high-pass filter defined by the lower limit of the heartbeat frequency
Implementation Method 3
projection data in which an animal is captured as a subject at each time by an X-ray CT apparatus
Data Source
AI summary
Provided are a CT-image processing apparatus and a method which can remove noise caused by one of a heartbeat and a breathing beat and extract only an accurate periodic motion by the other one. A CT-image processing apparatus 5 that processes projection data in which an animal is captured as a subject at each time by an X-ray CT apparatus includes a breathing beat threshold specification unit 36a to specify a lower limit of a breathing beat frequency from a feature amount waveform within a ROI for breathing beat synchronization in a series of projection data, a heartbeat threshold specification unit 37a to specify a lower limit of a heartbeat frequency from a feature amount waveform within a ROI for heartbeat synchronization in the series of projection data, a breathing beat extraction unit 36b to extract a breathing beat waveform using a band-pass filter defined by the lower limit of the breathing beat frequency and the lower limit of the heartbeat frequency.