CT Image Reconstruction for Bone Marrow Lesion Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computed tomography (CT) scans have limited sensitivity for detecting early bone marrow lesions, particularly in the axial skeleton, leading to missed diagnoses and unnecessary treatments, as they struggle to differentiate soft tissue components within bone structures, and are not suitable for patients with medical implants that react poorly to MRI.
Innovation Solution
A single-energy CT image processing routine that includes a tissue suppression component to identify and suppress voxels representing trabecular bone, allowing for voxel-by-voxel averaging to generate images with reduced axial resolution, and an attenuation transform component to map voxels to brightness values using a piecewise transform function, enhancing the visibility of bone marrow lesions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard CT scan reconstruction is used, then axial resolution is maintained, but sensitivity for detecting early bone marrow lesions is limited
Solution Approach 1:
The patent segments the CT image data into multiple thin-slab datasets along the axial direction. By dividing the volume into discrete slabs and processing them separately, the system can apply lesion-specific enhancement algorithms to each slab, improving sensitivity for detecting subtle bone marrow lesions while maintaining overall axial resolution through selective processing of relevant regions.
Solution Approach 2:
The patent applies different processing qualities to different regions of the image. Bone marrow regions receive enhanced processing with lesion-detection optimization, while other tissues maintain standard processing. This local differentiation allows improved lesion detection sensitivity without degrading overall image quality across the entire volume.
2Reliability
If voxel-by-voxel averaging is applied to reduce axial resolution, then image noise is reduced, but trabecular bone structures are obscured
Solution Approach 1:
The patent extracts trabecular bone structures from the image data using threshold-based segmentation and attenuation value analysis. By identifying and separating trabecular bone regions from bone marrow regions, the system can apply noise-reduction averaging to the bone marrow while preserving or removing trabecular bone contributions as needed, preventing obscuration of these important structural elements.
Solution Approach 2:
The patent changes processing parameters dynamically based on local tissue characteristics. In regions containing trabecular bone, the system adjusts averaging kernels and suppression thresholds to maintain structural visibility. In homogeneous bone marrow regions, full averaging is applied for noise reduction. This parameter adaptation resolves the contradiction between noise reduction and structural preservation.
3Measurement precision
If tissue suppression is applied to remove trabecular bone, then bone marrow lesion visibility is improved, but anatomical context is reduced
Solution Approach 1:
The patent implements dynamic, user-selectable suppression levels for trabecular bone. Rather than applying fixed suppression, the system allows adjustment of suppression thresholds and intensity, enabling users to balance lesion visibility against anatomical context preservation. This dynamic control resolves the contradiction by making the trade-off adjustable rather than fixed.
Solution Approach 2:
The patent preserves anatomical context by maintaining the full-resolution original images alongside the processed images with suppression. Users can reference the original anatomical context from unprocessed images while evaluating enhanced lesion visibility in processed images, effectively adding a temporal/dimensional dimension for context comparison rather than losing information permanently.
4Measurement precision
If piecewise attenuation transform is applied, then contrast between soft tissues is enhanced, but quantitative accuracy is reduced
Solution Approach 1:
The patent creates a transformed copy of the attenuation data with enhanced soft tissue contrast through piecewise linear transformation. This copied dataset is used specifically for visualization and qualitative assessment, while the original quantitative attenuation data is preserved for measurements and quantitative analysis. This copying approach allows enhanced contrast display without compromising the accuracy of quantitative attenuation values.
Data Source
AI summary
Systems and methods are provided for computer tomography (CT) imaging. An attenuation transform component configured to map voxels in a received set of cross-sectional CT images to associated brightness values according to a piecewise transform function to produce a set of transformed images. A user interface is configured to provide the set of transformed images to a user at an associated display.