Respiratory Model for Organ Tracking in HIFU Therapy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current HIFU treatment methods face challenges in accurately tracking the changing location and shape of organs during respiration, as diagnostic images are often imprecise due to patient movement and inability to maintain proper respiratory states, affecting the precision of tumor targeting.
Innovation Solution
A method and apparatus that generate a model representing changes in the location and shape of a region of interest during the respiration cycle using diagnostic images at two points in time, combined with 3D ultrasound image analysis for shape information extraction and registration, allowing for the generation of an updated model that accurately reflects tissue deformation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If diagnostic images are captured at full expiration and full inspiration to obtain organ location and shape changes, then the model accuracy is improved, but the patient cannot maintain proper respiratory states causing image quality degradation
Solution Approach 1:
A respiratory signal sensor is introduced as an intermediary to indirectly measure organ position changes. Instead of relying on patient cooperation to hold breath, the system uses the respiratory signal to drive a mathematical model that predicts organ location and shape, thereby eliminating the need for breath-holding while maintaining measurement accuracy.
Solution Approach 2:
The patent replaces the mechanical approach of capturing images at specific respiratory phases (which requires patient breath-holding) with a computational model-driven approach. The system substitutes direct mechanical imaging at controlled respiratory states with a mathematical model that calculates organ position based on respiratory signals, eliminating the need for mechanical breath-holding control.
2Ease of operation
If diagnostic images are captured requiring patient breath-holding, then the respiration state control is improved, but the treatment time increases due to patient inability to maintain respiratory states
Solution Approach 1:
The respiratory signal acts as an intermediary that continuously tracks respiratory phase without requiring the patient to consciously control or hold breath. This eliminates the time-consuming process of instructing and waiting for patients to achieve and maintain specific respiratory states, thereby reducing treatment time while maintaining respiration state control.
Solution Approach 2:
The system uses the patient's own natural respiratory signals to drive the model, eliminating the need for external control or patient effort to maintain breath-holding. The respiratory signal automatically provides the necessary information about respiratory phase, allowing the system to adapt to natural breathing patterns without additional time requirements.
3Measurement precision
If a mathematical model driven by respiratory signal is used to track organ position, then the treatment precision is improved, but the model complexity increases
Solution Approach 1:
The mathematical model is pre-calculated and established before treatment based on diagnostic images. This preliminary modeling phase captures the relationship between respiratory signals and organ position, allowing the treatment phase to simply query the pre-computed model rather than performing complex real-time calculations, thereby reducing treatment-time complexity while maintaining precision.
Solution Approach 2:
The patent creates a simplified computational copy or representation of the organ's position and shape as a function of respiratory phase. This model copy captures the essential geometric relationships without requiring full complexity of the actual organ structure, enabling efficient real-time tracking with reduced computational burden during treatment.
Data Source
AI summary
Provided is a method of generating a model, the method including generating a first model representing a change in the location or the shape of the region of interest during the respiration cycle, using diagnostic images that are obtained at two points of time in the respiration cycle and that represent the region of interest; extracting shape information of one or more tissues included in the region of interest at a shape information extractor, using a 3D ultrasound image that is obtained at one point of time in the respiration cycle; determining a characteristic point of the 3D ultrasound image corresponding to a characteristic point of the first model by matching the first model with the extracted shape information; and generating a second model by updating the first model with the determined characteristic point.