Operation Determination Model Training for Vascular Instrument Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medical procedures for inserting instruments like stents into blood vessels, such as cardiovascular and cerebrovascular interventions, pose challenges due to the physical burden on practitioners, including radiation exposure and the need for adapting to new tools for precise control.
Innovation Solution
A method and system for training an operation determination model using reinforcement learning and supervised learning to identify procedure environments in branching regions of vascular models, enabling accurate control of medical instruments by selecting appropriate reinforcement learning models based on vascular patch images and generating training data for the operation determination model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If a practitioner performs medical instrument control using visual and haptic information, then the procedure can be completed with basic tools, but the practitioner experiences physical burden including radiation exposure and difficulty in precise control
Solution Approach 1:
The patent replaces the mechanical system of manual instrument control with an automated control system that uses machine learning models (reinforcement learning and supervised learning) to determine operations. The control device automatically processes medical images and determines instrument operations, substituting the practitioner's manual control with an automated intelligent system that reduces radiation exposure while maintaining precise control capability.
2Object-affected harmful factors
If an automated control system is introduced to reduce practitioner burden, then radiation exposure is reduced, but the practitioner needs to learn to adapt to a new tool
Solution Approach 1:
The control system performs self-learning through reinforcement learning and supervised learning mechanisms. The system automatically trains operation determination models using medical images and operational data, enabling it to adapt to different procedure environments without requiring extensive practitioner retraining. The multi-model architecture with environment-specific models allows the system to serve itself by selecting appropriate models for different vascular branching scenarios.
3Measurement precision
If multiple reinforcement learning models are used for different procedure environments, then the accuracy of operation determination is improved, but the device complexity increases
Solution Approach 1:
The patent segments the procedure environment into distinct categories based on vascular branching characteristics. Multiple reinforcement learning models are trained for different environment types (e.g., different branching patterns, vessel diameters). The system identifies the current environment type and selects the corresponding specialized model, allowing each model to be optimized for its specific environment while managing complexity through structured segmentation.
Solution Approach 2:
The system performs preliminary training of multiple reinforcement learning models for different procedure environments before actual use. Environment-specific models are pre-trained and stored, allowing rapid selection during procedures without real-time complex computations. The preliminary preparation of environment classifications and model selections reduces runtime complexity while maintaining high accuracy.
Data Source
AI summary
This system for training an operation determination model for a medical instrument control device generates training data by means of a reinforcement learning model, and, using the training data, can train an operation determination model configured to output information associated with operational commands for a driving unit that transports medical instruments.