Truncation Correction in Fan Beam CT via Body Contour Extrapolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Transmission CT apparatuses using fan beam collimators face truncation issues when imaging larger objects, leading to incomplete reconstruction and inaccurate image acquisition due to a limited effective acquisition field, which cannot be effectively addressed by existing methods without hardware changes or low detection accuracy from emission CT apparatuses.
Innovation Solution
A radiodiagnostic apparatus that collects first and second projection data within and outside the effective acquisition field, uses primary and secondary correction units to extrapolate and correct truncated data, and detects body contours to reconstruct images accurately, allowing for high-accuracy attenuation distribution measurement and truncation correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a fan beam collimator is used to reduce scattered radiation and improve attenuation coefficient measurement accuracy, then measurement precision is improved, but the effective acquisition field is narrowed causing truncation artifacts
Solution Approach 1:
The patent segments the projection angle range into two distinct parts: a first projection angle range where the entire object is within the effective acquisition field, and a second projection angle range where part of the object is truncated. This segmentation allows independent processing of complete and truncated data, enabling accurate body contour detection from the complete portion while separately handling the truncated portion through extrapolation techniques.
Solution Approach 2:
The patent performs preliminary actions by first collecting complete projection data from the first projection angle range before dealing with the truncated second projection angle range. The body contour is detected from the complete data, and this contour information is then used to guide the extrapolation and correction of the truncated data, ensuring that the correction process is based on accurate anatomical boundaries.
2Reliability
If the effective acquisition field is enlarged to accommodate larger objects and eliminate truncation, then truncation artifacts are reduced, but the apparatus becomes unnecessarily enlarged and cost increases
Solution Approach 1:
The patent replaces the mechanical solution of enlarging the physical acquisition field with a computational approach. By using software-based extrapolation techniques that leverage the body contour detected from complete projection data, the system can accurately reconstruct truncated regions without requiring a larger detector or collimator, thus avoiding apparatus enlargement and associated cost increases.
Solution Approach 2:
The patent changes the parameter space by working with projection angle ranges rather than physical field size. By identifying and utilizing the first projection angle range where complete object coverage is achieved, the system can extract accurate body contour information that remains valid even when the object extends beyond the effective acquisition field in other angles, enabling accurate reconstruction without hardware changes.
3Measurement precision
If body contour is detected from emission CT apparatus to solve truncation, then truncation correction can be made, but detection accuracy is low and requires additional emission CT hardware
Solution Approach 1:
The patent makes the transmission CT apparatus multi-functional by enabling it to perform both attenuation distribution measurement and body contour detection using the same hardware system. The transmission CT data, originally intended only for attenuation measurement, is utilized to detect the body contour from the first projection angle range, eliminating the need for separate emission CT hardware and improving contour detection accuracy through the high-resolution transmission data.
Solution Approach 2:
The transmission CT apparatus serves itself by using its own collected projection data to detect the body contour and perform truncation correction. The complete projection data from the first projection angle range is processed to extract body contour information, which then feeds back into the reconstruction process to correct truncation artifacts in the second projection angle range, making the system self-sufficient without requiring external emission CT equipment.
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
Enables accurate measurement of attenuation distribution and truncation correction, as well as high-accuracy detection of body contours directly from a transmission CT apparatus, reducing errors and improving image reconstruction quality.
Implementation Method 1
measuring the radiation having transmitted from the object
Implementation Method 2
a detector as a pair around an axis of the object and measuring the radiation having transmitted from the object
Implementation Method 3
a transmission CT apparatus using a fan beam collimator has less scattered radiation since gamma rays are collimated at two points, at the radiation source and at the collimator
Implementation Method 4
rotating a radiation source of radiation (X-ray or gamma ray) installed outside the object and a detector as a pair around an axis of the object
Implementation Method 5
obtain a tomographic distribution of radiographic transmission of an object
Data Source
AI summary
A radiodiagnostic apparatus includes a data collection unit, a primary correction unit, a body contour detection unit, a secondary correction unit, and a reconstruction unit. The primary correction unit calculates a plurality of primary extrapolation expression candidates for correcting second projection data so as to obtain third projection data candidates formed from the second projection data and a plurality of primary extrapolation expression candidates and for obtaining third projection data from the plurality of third projection data candidates.