AI Puncture Path Planning for Vertebral Pedicle Safety
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for planning puncture paths in PVP surgery lack objectivity and accuracy, leading to potential bone cement leakage and associated complications, as they rely heavily on manual experience and subjective judgment.
Innovation Solution
An intelligent planning method using artificial intelligence algorithms, specifically a trained neural network model, to identify optimal puncture injection stop-points and calculate comprehensive distances to determine the safest and most effective puncture paths, minimizing bone cement leakage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual puncture path judgment through intraoperative X-ray fluoroscopy is used, then the surgery can be performed with existing equipment, but the planning is highly subjective and lacks objectivity
Solution Approach 1:
The patent replaces manual mechanical judgment (doctor's experience-based visual assessment) with an automated computer-aided system that uses AI algorithms to objectively predict key point positions and generate puncture paths, thereby improving measurement precision while maintaining reasonable device complexity
Solution Approach 2:
The system enables the computer to automatically perform the planning task that previously required human expertise, with the AI model self-identifying key anatomical points and generating optimal puncture paths without continuous human intervention
2Ease of manufacture
If traditional empirical puncture path for spinal screw implantation is used, then the method is simple to implement, but it is not suitable for percutaneous puncture of the vertebral body in PVP surgery and it is difficult to ensure that bone cement does not leak
Solution Approach 1:
The patent changes the planning parameters specifically for PVP surgery by identifying stop points at the anterior 1/3 of the vertebral body rather than using traditional spinal screw parameters (center point at 1/3 middle and anterior), thereby improving reliability for bone cement injection while adapting the approach to PVP-specific requirements
Solution Approach 2:
The system applies different planning strategies for different surgical procedures by detecting the specific surgery type (PVP vs. spinal screw implantation) and adjusting key point positions accordingly, with PVP using anterior 1/3 stop points and spinal screw surgery using center point positioning
3Measurement precision
If computer-aided guidance planning is used, then objective planning can be achieved, but it is difficult to obtain relevant clinical data on bone cement leakage, so it is impossible to establish a judgment model about bone cement leakage in a short time
Solution Approach 1:
The patent performs preliminary action by pre-training the AI model on spinal CT images to learn anatomical structures and key point positions before actual PVP surgery, allowing the system to be ready for immediate use without requiring time-consuming model establishment during the surgical procedure
Solution Approach 2:
The system pre-establishes constraint strategies for bone cement injection stop-points by training the AI model in advance on clinical data and anatomical variations, so that when PVP surgery is performed, the model is already configured to prevent bone cement leakage without requiring additional time for model development
Data Source
AI summary
An intelligent puncture path planning method, a device, an equipment and a medium in a PVP operation. The method includes: intercepting a local CT image of a vertebral body from an image; identifying midpoint of the narrowest part of vertebral pedicle and a theoretical puncture injection stop-point from the local CT image through trained neural network model; calculating distance from the stop-point to the middle plane in front-back direction of the vertebral body, distance from the stop-point to the middle plane in left-right direction of the vertebral body and the closest distance from the stop-point to the cortical bone surface, taking the minimum value of distances as radius of a space sphere area with the stop-point as sphere center; taking a line connecting midpoint of the narrowest part of any vertebral pedicle and any point in the space sphere area as an optional puncture path of the vertebral pedicle.