Automated HIFU Treated Volume Segmentation in MR Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for segmenting HIFU-ablated tissue from surrounding tissue in MR images are manual, time-consuming, and lack accuracy, necessitating a more efficient and automated approach for volume estimation and visualization.
Innovation Solution
A computer-implemented method and medical imaging system that uses contrast-enhanced T1 weighted magnetic resonance images to identify treated volumes by segmenting medical imagery based on contrast agent release, enabling automated or semi-automatic segmentation and visualization of HIFU-ablated tissue, with seed points derived from planned treating locations and thermal models for precise volume calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual contouring of ablation edges is performed slice-by-slice from contrast-enhanced MR images, then segmentation accuracy can be achieved, but the process becomes time-consuming and labor-intensive
Solution Approach 1:
The system enables automated segmentation where the computer system automatically identifies and contours ablation edges in MR images without requiring manual slice-by-slice contouring. The automated algorithm processes the contrast-enhanced images to delineate treated volumes, making the system self-sufficient and eliminating the need for time-consuming manual intervention while maintaining segmentation accuracy.
Solution Approach 2:
The manual mechanical process of slice-by-slice contouring is replaced with an automated computational algorithm. The computer system uses image processing and pattern recognition techniques to automatically identify ablation edges in contrast-enhanced MR images, substituting the manual mechanical contouring process with an automated digital solution that significantly reduces processing time while maintaining or improving accuracy.
2Productivity
If automated segmentation algorithms are implemented, then processing speed and efficiency are improved, but accuracy and reliability of treated volume identification may be compromised
Solution Approach 1:
The system incorporates feedback mechanisms where the automated segmentation algorithm processes contrast-enhanced MR images and generates treated volume identifications that can be evaluated and refined. The feedback loop allows the system to learn from results and adjust parameters to maintain high accuracy while operating at automated speeds, ensuring reliability is not compromised by the increase in processing speed.
Solution Approach 2:
The segmentation approach combines multiple techniques and data sources - including contrast-enhanced MR imaging data, automated image processing algorithms, and potentially manual verification components - into a composite segmentation system. This multi-component approach leverages the strengths of each element to achieve both high processing speed and maintained accuracy, creating a reliable automated solution that exceeds the capabilities of single-method approaches.
3Loss of information
If contrast agents are administered after therapy to evaluate treatment outcome, then treated volume identification is enabled, but the timing and coordination of imaging with therapy delivery becomes complex
Solution Approach 1:
The medical imaging system is designed to perform multiple functions - it serves both as a therapy delivery system during HIFU treatment and as an evaluation system after therapy using contrast-enhanced MR imaging. This multi-functional capability allows the same imaging infrastructure to be used for real-time therapy guidance and post-treatment outcome assessment, reducing overall system complexity despite the coordinated timing requirements.
Solution Approach 2:
The system performs preliminary actions by establishing the imaging infrastructure and protocols before therapy delivery. Contrast agent administration timing and imaging parameters are pre-planned and coordinated with the therapy protocol, so that when therapy is delivered and subsequently evaluated, the imaging system is already configured and ready to capture the treatment outcome data without adding complexity during the actual treatment execution.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Facilitates fast, accurate, and easy-to-use segmentation of treated volumes, allowing for real-time feedback and adaptive therapy adjustments, thereby enhancing the effectiveness and precision of tissue treatment outcomes.
Implementation Method 1
gadolinium based contrast agents, such as Gd-DTPA, gadodiamide, or gadoteridol, affect the T1 relaxation time and may be used to identify non-perfused volumes within the subject
Implementation Method 2
high intensity focused ultrasound is used for sonicating or treating a region of tissue within a subject with high intensity focused ultrasound. The high intensity focused ultrasound can be used for heating a region within the subject
Implementation Method 3
at higher powers cavitation can be used to destroy regions within the subject
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A medical imaging system (900, 1000, 1100, 1200) for acquiring medical image data (930), the medical imaging system comprising: a tissue treating system (910, 1080, 1180, 1190, 1280, 1290) for treating a target volume (908); a computer system (918) comprising a processor (922), wherein the computer system is adapted for controlling the medical imaging system; and a memory (928) containing machine readable instructions (954, 956, 958, 962, 964, 966, 968, 970, 972, 974). Execution of the instructions cause the processor to: acquire (100, 200, 308) medical image data; reconstruct (102, 202, 310) a medical image (932) using the medical image data; receive (104, 204, 312) an image segmentation seed (600, 934) derived from a treatment plan (936), and identify (106, 210, 314) a treated volume (400, 700, 800) in the medical image by segmenting the medical image in accordance with the image segmentation seed.