Hierarchical Tumor Motion Estimation for Whole-Organ Respiratory Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for precisely inserting a puncture needle into a tumor position are hindered by tumor and organ motion due to patient respiration, leading to inaccurate positioning of tumors and surrounding vital structures.
Innovation Solution
A hierarchical motion estimation method using a tumor motion estimation model and an organ motion estimation model, constructed with priori knowledge from pre-collected image data, to estimate respiratory motion of the whole organ in real time, improving positioning accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a respiratory motion model is established using correlation between proxy signal and internal target motion, then real-time target motion estimation is achieved, but the method can only simulate respiratory motion of a single anatomical feature point and is not applicable to a whole organ
Solution Approach 1:
The patent segments the motion estimation problem into two hierarchical levels: tumor-level motion estimation using proxy signal correlation, and organ-level motion estimation using velocity vector field propagation. This segmentation allows the system to first accurately track tumor motion and then extend this to whole-organ motion, resolving the contradiction between single-point accuracy and whole-organ applicability
Solution Approach 2:
The patent implements a nested estimation architecture where the tumor motion estimation model is embedded within the organ motion estimation model. The tumor position serves as input to the organ-level model, which propagates motion to surrounding anatomical structures. This nesting enables the system to maintain accurate tumor positioning while simultaneously providing whole-organ motion estimation
2Device complexity
If a respiratory motion model is established for single point motion, then modeling complexity is reduced, but accurate positioning of tumor and surrounding vital anatomical structures cannot be achieved
Solution Approach 1:
The patent performs preliminary action by pre-calculating and storing velocity vector fields for the organ in different respiratory states before actual motion estimation. These pre-computed velocity fields are then used during real-time operation to rapidly estimate organ motion without complex on-the-fly calculations, thereby maintaining low complexity while achieving high positioning accuracy for both tumor and surrounding structures
Solution Approach 2:
The patent introduces velocity vector fields as an intermediary between tumor motion estimation and organ motion estimation. The velocity vector fields act as a mediator that translates tumor position information into comprehensive organ-level motion, enabling accurate positioning of both the tumor and surrounding vital anatomical structures without requiring direct complex modeling of all structures
Data Source
AI summary
The present disclosure relates to the technical field of image processing and provides a motion estimation method and apparatus for a tumor, a terminal device, and a storage medium. The method includes: obtaining a current value of a respiratory correlation signal of a patient, the respiratory correlation signal reflecting a motion trajectory feature of a designated organ with the tumor of the patient in each of different respiratory states; inputting the current value of the respiratory correlation signal into a tumor motion estimation model to obtain an estimated current position of the tumor; and inputting the estimated current position of the tumor into an organ motion estimation model to obtain an estimated current velocity vector field of the designated organ. By adopting the motion estimation method, it realizes a respiratory motion estimation for the whole organ and enhance accuracy of positioning the tumor and vital anatomical structures around the tumor.