Neural Network Medical Trajectory Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical intervention planning techniques are manual, incomplete, and imprecise, failing to account for anatomical deformation and requiring operator expertise, leading to errors in trajectory planning for medical instruments.
Innovation Solution
An automatic trajectory planning method using a neural network trained on medical images to determine optimal entry points and trajectories, independent of operator intervention, which can handle non-segmented images and account for anatomical constraints and instrument deformation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual planning by operator is used, then operator expertise and judgment can be applied, but the process is tedious, restrictive, and requires significant attention and experience
Solution Approach 1:
The system performs trajectory planning autonomously by processing medical images and generating optimal trajectories without requiring continuous operator intervention. The automated system analyzes anatomical structures, identifies entry points, and calculates trajectories that satisfy constraints, making the planning process self-service and reducing dependency on operator expertise.
Solution Approach 2:
The patent replaces manual mechanical planning with an automated computational system that uses image processing algorithms and mathematical models to calculate optimal trajectories. This substitution eliminates the need for manual measurement and visual estimation, replacing operator-based mechanics with automated computational mechanics.
2Measurement precision
If classical image processing algorithms with segmentation are used, then trajectories can be determined based on predefined constraints, but segmentation proves imprecise and incomplete leading to errors in trajectory
Solution Approach 1:
The system changes the fundamental parameters of image processing by using deep learning neural networks instead of traditional segmentation algorithms. The neural network directly processes medical images to identify anatomical structures and calculate optimal trajectories, avoiding the imprecise segmentation step while improving both measurement precision and reliability through learned features from training data.
3Adaptability or versatility
If predefined constraints and scoring systems are used, then multiple entry points can be proposed with scores, but the operator must still manually select and verify trajectories requiring significant expertise
Solution Approach 1:
The automated system independently performs trajectory selection by evaluating multiple possible trajectories against predefined constraints and selecting the optimal one without operator intervention. The system self-verifies trajectory validity by checking anatomical constraints, ensuring no passage through critical structures, and confirming trajectory feasibility, thereby eliminating the need for manual verification while maintaining adaptability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for automatically planning a trajectory to be followed during a medical intervention by a medical instrument (120) targeting an anatomy of interest (130) of a patient (110), said automatic planning method comprising the steps of: - acquiring at least one medical image of the anatomy of interest (130); - determining a target point (145) on the previously acquired image; - generating a set of trajectory planning parameters from the medical image of the anatomy of interest and the previously determined target point, the set of planning parameters comprising coordinates of an entry point on the medical image. The set of parameters is generated using a machine learning method of neural network type. The invention also relates to a guiding device (150) implementing the set of planning parameters obtained.