Tomosynthesis Image Reconstruction Using Native Projection Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Tomosynthesis imaging systems face challenges with artefacts caused by biopsy needles, such as streak artefacts and limited field-of-view due to truncated projections, which affect the accuracy of verification images during guided biopsies.
Innovation Solution
The solution involves using native image data to augment verification images by adding or replacing projection data, thereby enlarging the field of view and removing truncation and high-contrast artefacts. This is achieved by selecting and processing subsets of projection data from the native image to generate synthetic forward projections, which are then used in conjunction with verification image data to reconstruct enhanced images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If verification images are acquired during tomosynthesis guided biopsies, then accurate 3D information about needle location is obtained, but streak artefacts are generated due to the presence of the high contrast needle
Solution Approach 1:
The patent applies preliminary action by acquiring native projection data before the biopsy needle is inserted. This pre-acquired data is then combined with verification projection data to reconstruct images that are free from needle-induced streak artefacts, while still providing accurate 3D needle location information.
Solution Approach 2:
The patent uses native projection data as an intermediary to mediate between the verification images and the final artefact-free reconstruction. The native data serves as a substitute for the corrupted verification data, allowing reconstruction of images without streak artefacts while maintaining diagnostic accuracy.
2Measurement precision
If dedicated acquisition with higher dose and/or higher tomographic angle is performed for boosted tomosynthesis, then diagnostic conspicuity of sub-region structures is improved, but field-of-view is limited due to truncated projections
Solution Approach 1:
The patent merges native projection data (acquired over a wide field-of-view) with boosted projection data (acquired with higher dose and angle for diagnostic conspicuity). By combining these datasets, the reconstruction achieves both high diagnostic quality in the sub-region and complete field-of-view coverage without truncation artefacts.
Data Source
Figure 1~2
Figure 3a~4
AI summary
The present invention relates to an apparatus for tomosynthesis image reconstruction. It is described to provide (210) a first projection image data set and a second projection image data set acquired after acquisition of the first projection image data set, wherein, the first projection image data set comprises first projection data, and the second projection image data set comprises second projection data, and wherein the first projection image data set is useable for the reconstruction of a tomosynthesis image of at least a region of interest of a body part and wherein the second projection image data set is useable for the reconstruction of a tomosynthesis image of at least the region of interest of the body part. A subset of the first projection data is selected (220). An enhanced image of at least the region of interest of the body part is reconstructed (230) on the basis of the subset of the first projection data and second projection reconstruction data comprising at least a first subset of the second projection data. Information is output (240) relating to the enhanced image of at least the region of interest of the body part.