Digital Breast Tomosynthesis Lesion Segmentation for Precise Biopsy Targeting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital breast tomosynthesis systems face challenges in accurately positioning biopsy needles due to incorrect needle placements, which can be disguised by overlapping tissues and vessels, and lack effective methods for outlining lesions for precise targeting.
Innovation Solution
A computer-implemented method that uses a trained function, such as a segmentation or region growing method, to generate an outline of a lesion in digital breast tomosynthesis by applying a trained function to a reconstructed tomosynthesis volume dataset, incorporating a virtual target marker to establish a two-dimensional or three-dimensional outline of the lesion, which is then displayed alongside projection recordings for precise biopsy targeting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional tomosynthesis imaging is used for biopsy targeting, then the imaging workflow can be maintained, but incorrect needle placements occur due to overlapping tissues and vessels disguising the lesion
Solution Approach 1:
The patent applies segmentation by separating the lesion from surrounding tissues through 3D volume rendering and multi-planar reconstruction. The lesion is segmented as a distinct volumetric object with defined boundaries, allowing it to be visualized independently from overlapping tissues and vessels in the tomosynthesis data set.
Solution Approach 2:
The patent transitions from 2D planar imaging to 3D volumetric visualization to resolve tissue overlap. By rendering the lesion as a three-dimensional object with spatial coordinates (x, y, z), the system provides depth information that eliminates the harmful effect of superimposed structures in conventional 2D views.
2Measurement precision
If AI-based CAD methods are used for lesion detection, then lesion probability maps can be generated, but the methods lack precise outlining capability for biopsy targeting
Solution Approach 1:
The patent performs preliminary outlining of the lesion volume before biopsy needle insertion. The system calculates the 3D lesion boundaries and generates visual outlines in advance, allowing the operator to plan the biopsy trajectory with precise knowledge of the lesion extent before the invasive procedure begins.
Solution Approach 2:
The patent introduces an intermediary visualization layer between the raw tomosynthesis data and the biopsy targeting decision. This intermediary consists of computed 3D lesion outlines and probability maps that mediate between the complex imaging data and the simple binary decision of whether to proceed with biopsy at a specific location.
3Difficulty of detecting and measuring
If multiple synthetic mammograms are generated for rotating visualization, then lesion visibility is improved, but the processing time and computational load increase
Solution Approach 1:
The patent applies partial action by generating and displaying only the most relevant synthetic mammogram views for biopsy targeting, rather than computing all possible rotating views. The system selectively renders synthetic images at specific angles that optimize lesion visibility and needle path planning, reducing unnecessary computational effort.
Data Source
AI summary
One or more example embodiments of the present invention relates to a computer-implemented method for providing an outline of a lesion in digital breast tomosynthesis includes receiving input data, wherein the input data comprises a reconstructed tomosynthesis volume dataset based on projection recordings, a virtual target marker within a lesion being in the tomosynthesis volume dataset; applying a trained function to at least a part of the tomosynthesis volume dataset to establish an outline enclosing the lesion, the part of the tomosynthesis volume dataset corresponding to a region surrounding the virtual target marker in the tomosynthesis volume dataset; and providing output data, wherein the output data is an outline of a two-dimensional area or a three-dimensional volume surrounding the target marker.