Electrode Position Signals with Patient-Specific Model Retraining
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current electro-anatomical mapping systems face challenges in accurately detecting magnetic and impedance signals across different subjects due to variations between subjects, leading to inaccurate impedance and cardiac calibrations.
Innovation Solution
A system and method using a retrained machine-learning model that includes a catheter with electrodes to collect potential signals and a magnetic sensor to collect magnetic data, combined with a computing device to receive patient-agnostic and patient-specific data sets, train and retrain a mapping machine-learning model, and generate precise position signals for electrodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a fixed mapping machine-learning model is used for electrode positioning, then the system structure remains simple, but measurement precision deteriorates due to inability to adapt to patient variations
Solution Approach 1:
The system performs preliminary actions by training an initial mapping machine-learning model using patient-agnostic data before actual use. This pre-trained model provides a baseline capability for electrode positioning that can then be refined. The preliminary training establishes general anatomical relationships that work across multiple patients, creating a foundation that improves measurement precision without requiring complete model retraining for each patient.
Solution Approach 2:
The system implements dynamics by enabling the mapping machine-learning model to adapt and evolve from a static patient-agnostic model to a dynamic patient-specific model. The model transitions from fixed parameters to adaptable parameters through retraining with patient-specific calibration data, allowing it to adjust to individual anatomical variations while maintaining the underlying framework of the original model.
2Measurement precision
If patient-specific calibration is performed for each subject, then measurement precision improves, but loss of time increases due to additional calibration procedures
Solution Approach 1:
The system performs preliminary impedance and cardiac calibrations during the initial patient-agnostic training phase, establishing baseline calibration parameters before patient-specific procedures. This preliminary calibration captures general anatomical and electrical characteristics, reducing the amount of calibration needed during actual patient procedures and thereby minimizing time loss while maintaining precision.
Solution Approach 2:
The system implements feedback mechanisms where patient-specific calibration results are used to refine and update the mapping machine-learning model. The calibration process provides feedback about individual patient characteristics, which are then fed back into the model to improve subsequent measurements. This feedback loop enables accurate patient-specific calibration without requiring extensive manual adjustment time, as the automated feedback-driven refinement streamlines the calibration process.
3Adaptability or versatility
If a retrained machine-learning model is implemented to adapt to patient variations, then adaptability improves, but device complexity increases due to dual training requirements
Solution Approach 1:
The system applies segmentation by dividing the training data into distinct segments: patient-agnostic training data that establishes general anatomical relationships, and patient-specific calibration data that captures individual variations. This segmentation allows the model to process different types of information separately and combine them effectively, improving adaptability while managing complexity through structured data organization and separate processing pipelines for each training segment.
Data Source
AI summary
A system for determining position signals of electrodes using a retrained machine-learning model includes at least a catheter including a plurality of electrodes configured to collect a plurality of potential signals and a magnetic sensor configured to collect magnetic data, and at least a computing device including a memory. The processor receives a first training set, wherein the first training set includes patient-agnostic data; receives a second training set, wherein the second training set includes patient-specific data, trains a mapping machine-learning model using the first training set, retrains the mapping machine-learning model using the second training set, receives at least a first signal, wherein the first signal includes a potential signal of the plurality of potential signal and the magnetic data, and generates, using the retrained machine-learning model, as a function of the at least a first signal, a first position signal for an electrode of the plurality of electrodes.