4D Cone Beam CT Projection Allocation for Streak Artefact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Four-dimensional cone beam computed tomography (4DCBCT) images suffer from streak artefacts due to constant gantry speed and projection pulse rate, which degrade image quality and make it challenging to accurately position tumors during radiotherapy.
Innovation Solution
Optimizing respiratory bin position, size, and projection allocation by reducing the variance of angular separation between projections, sharing projections between adjacent bins, and using a simple heuristic to minimize standard deviation, thereby improving image quality and reducing streak artefacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If constant gantry speed and projection pulse rate are used, then acquisition process is simple, but streak artefacts occur and image quality degrades
Solution Approach 1:
The patent applies dynamics by transitioning from constant gantry speed to variable gantry speed that adapts to respiratory motion phases. The system dynamically adjusts the gantry rotation speed to be slower during phases with significant respiratory motion and faster during stable phases, thereby reducing streak artifacts while maintaining acquisition efficiency.
Solution Approach 2:
The patent implements parameter changes by modifying the gantry speed parameter based on respiratory phase detection. The system changes the gantry rotation speed parameter dynamically during the acquisition process, adjusting it according to the detected respiratory motion to optimize image quality and minimize artifacts.
2Manufacturing precision
If projections are densely sampled, then image quality improves, but acquisition time increases
Solution Approach 1:
The patent applies local quality by differentiating the sampling density across different respiratory phases. Instead of uniform dense sampling throughout the entire acquisition, the system applies higher projection sampling density locally during stable respiratory phases where it contributes most to image quality, and reduces sampling during phases with significant motion.
Solution Approach 2:
The patent implements partial action by selectively applying dense projection sampling only during beneficial respiratory phases rather than continuously. The system performs excessive sampling (dense projections) only when it provides value during stable phases, and reduces or skips sampling during unstable phases, optimizing the balance between image quality and acquisition time.
3Device complexity
If respiratory bin size is increased, then fewer bins are needed and processing is simpler, but angular separation variance increases and streak artefacts worsen
Solution Approach 1:
The patent applies dynamics by making the respiratory bin size adaptive rather than fixed. The system dynamically adjusts bin boundaries and sizes based on the detected respiratory signal characteristics, ensuring that each bin contains projections with relatively uniform angular separation while adapting to the actual respiratory motion patterns observed during acquisition.
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
The proposed method significantly reduces streak artefacts and improves image quality in 4DCBCT images, allowing for more accurate tumor positioning and potentially reducing imaging time and dose, with a near-optimal solution achievable through heuristic methods.
Implementation Method 1
a source of X-rays is rotated about a patient
Implementation Method 2
Four dimensional cone beam computed tomography (4DCBCT) imaging
Data Source
Figure 1
Figure 2~4
Figure 5
AI summary
A method, and a system when implementing a method, of reducing artefacts in image creation in 4D cone beam computed tomography (4DCBCT) images, the method comprising the steps of: (a) performing a 4DCBCT scan of a target patient including a series of spaced apart projections through the target patient, with each projection having an associated estimated or measured respiratory state; (b) initially dividing the series of projections into a corresponding series of respiratory bins, with each respiratory bin having projections substantially from a portion of a cyclic respiratory state; and (c) optimising the projections at the bounds of each respiratory bin so as to improve an image quality measure of the images.