Real-Time Organ Segmentation for Catheter Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In HDR brachytherapy, the accuracy of catheter insertion near the prostate's periphery is hindered by the subjective and manual delineation of prostate boundaries from TRUS images, leading to potential bladder punctures and uneven catheter distribution, which can increase radiation exposure to normal tissues.
Innovation Solution
An interventional therapy system that includes a controller for segmenting reference image datasets to determine peripheral outlines, deformably registering these outlines with real-time ultrasound images, and optimizing catheter placement based on threshold boundary changes, ensuring accurate positioning of catheters relative to the prostate's boundaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual delineation of prostate boundaries from TRUS images is used, then the process is simple and quick, but the accuracy of catheter placement deteriorates
Solution Approach 1:
The patent replaces the manual mechanical delineation process with an automated image processing system that uses ultrasound image analysis and deformable registration algorithms to automatically determine prostate boundaries and provide real-time catheter placement guidance, thereby improving accuracy while reducing operational complexity
Solution Approach 2:
The patent introduces a computer-based image processing system as an intermediary between the ultrasound imaging and catheter placement processes. This intermediary automatically analyzes images, delineates boundaries, and provides guidance feedback, eliminating the need for manual boundary drawing while enhancing placement precision
2Measurement precision
If real-time image processing and deformable registration are implemented, then catheter placement accuracy is improved, but processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary segmentation and boundary identification on reference images before the actual catheter insertion procedure. This pre-processing allows the system to have boundary information ready in advance, reducing the computational burden during real-time guidance and minimizing processing delays
Solution Approach 2:
The patent employs deformable registration that can dynamically adapt to tissue deformation during catheter insertion. The system updates boundary information in real-time based on actual tissue movement, maintaining accuracy throughout the procedure without requiring complete re-processing of all images
3Object-affected harmful factors
If catheters are inserted closer to prostate boundaries to improve dose coverage, then radiation exposure to normal tissues is reduced, but the risk of bladder puncture increases
Solution Approach 1:
The patent implements real-time feedback through continuous ultrasound imaging and automated boundary tracking during catheter insertion. The system provides immediate visual feedback showing the catheter's position relative to the prostate boundary, allowing the operator to adjust the insertion path to maintain optimal distance from the boundary, thereby reducing radiation exposure to normal tissues while preventing bladder puncture
Solution Approach 2:
The patent uses pre-procedure imaging and boundary mapping to identify potential risk areas before catheter insertion. The system creates a planned insertion path that anticipates and avoids bladder puncture risks while optimizing catheter placement near prostate boundaries for improved dose coverage
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
An interventional therapy system (100, 200, 300, 900) may include at least one controller (102, 202, 910) which may obtain a reference image dataset (540) of an object of interest (OOI); segment the reference image dataset to determine peripheral outlines (545) of the OOI in the plurality image slices; acquire a current image of the OOI (548) using an ultrasound probe (114, 224); select a peripheral outline (CBS, 545) of a selected image slice of the plurality of slices of the reference image dataset which is determined to correspond to the current image; and/or modify the selected peripheral outline of the image slice of the plurality of slices of the reference image dataset in accordance with at least one deformation vector (549).