Iterative Image Reconstruction for CT Artifact Reduction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current CT image reconstruction algorithms, particularly analytical methods, face limitations such as assuming noise-free data, leading to image artifacts and suboptimal noise characteristics, and are restrictive in data acquisition protocols and detector geometry, while iterative algorithms improve noise performance but are computationally complex and require iterative refinements.

Innovation Solution

The method involves converting measured projection data into image data, regularizing it to generate synthesized projection data, isolating inconsistent regions, correcting the data, and weighting portions based on inconsistency to refine the image reconstruction process, allowing for improved noise handling and flexibility in data acquisition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If analytical image reconstruction algorithms are used, then reconstruction speed is improved, but image quality deteriorates due to artifacts and suboptimal noise characteristics

Engineering Contradiction:

Improvereconstruction speedVSAvoidimage quality

Core Design Contradiction:

SpeedVSManufacturing precision

Solution Approach 1:

The patent introduces an intermediary iterative refinement process between the analytical reconstruction and the final image output. The analytical algorithm provides an initial estimate, which then undergoes iterative optimization using measured projection data to eliminate artifacts and improve noise characteristics, achieving both speed and quality.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The analytical reconstruction algorithm performs a preliminary action by generating an initial image estimate quickly. This preliminary result is then refined through subsequent iterative steps that correct artifacts and optimize noise properties, combining the speed advantage with quality improvement.

Inventive Principle:

Principle #10Preliminary action

2Ease of manufacture

If analytical image reconstruction algorithms are used, then ease of implementation is improved, but adaptability deteriorates due to restrictive data acquisition protocols and detector geometry constraints

Engineering Contradiction:

Improveease of implementationVSAvoidadaptability to data acquisition protocols

Core Design Contradiction:

Ease of manufactureVSAdaptability or versatility

Solution Approach 1:

The patent transitions from a static analytical approach to a dynamic iterative process. The reconstruction algorithm adapts dynamically to different data acquisition protocols and detector geometries by iteratively adjusting the image estimate based on the specific characteristics of the measured projection data, enabling versatility while maintaining implementation feasibility.

Inventive Principle:

Principle #15Dynamics

3Manufacturing precision

If iterative image reconstruction algorithms are used, then image quality is improved, but computational complexity increases

Engineering Contradiction:

Improveimage qualityVSAvoidcomputational complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The analytical reconstruction algorithm performs a preliminary action by generating an initial image estimate that is already reasonably accurate. This preliminary result reduces the number of iterative refinement steps needed, thereby lowering the overall computational complexity while maintaining the quality benefits of iterative methods.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent employs a hybrid approach that skips the computationally intensive initial stages of pure iterative reconstruction by using the fast analytical method first. This allows the system to rush through the early reconstruction phases quickly and then apply iterative refinement only where needed to achieve final quality goals.

Inventive Principle:

Principle #21Skipping (Rushing through)

4Loss of time

If analytical image reconstruction algorithms are used, then processing time is reduced, but reliability deteriorates due to assumptions of noise-free data and ideal line integrals

Engineering Contradiction:

Improveprocessing timeVSAvoidreliability of reconstruction

Core Design Contradiction:

Loss of timeVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where the analytical reconstruction result is evaluated against the actual measured projection data, and discrepancies are used to guide iterative refinements. This feedback loop eliminates the need for unrealistic assumptions about noise-free data by continuously adjusting the reconstruction to match real measurements, improving reliability without excessive time cost.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS9824467B2Iterative image reconstruction

Publication Date: 2017.11.21 ANALOGIC CORP

AI summary

One or more techniques and/or apparatuses described herein provide for reconstructing image data of an object under examination from measured projection data indicative of the object. The measured projection data is converted into image data using an iterative image reconstruction approach. The iterative image reconstruction approach may comprise, among other things, regularizing the image data to adjust a specified quality metric of the image data, identifying regions of the image data that represent aspects of the object that might generate inconsistencies in the measured projection data and correcting the measured projection data based upon such an identification, and/or weighting projections comprised in the measured projection data differently to reduce the influence of projections that respectively have a higher degree of inconsistency in the conversion from projection data to image data.