Neural Network CPM Matrix Generation for Cardiac Mapping

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for generating comprehensive current-to-position (CPM) matrices in cardiac mapping are time and labor-intensive, requiring extensive data acquisition across various locations, which hinders efficient visualization and treatment of cardiac arrhythmias like atrial fibrillation.

Innovation Solution

A system and method utilizing a learning system, such as a neural network, to generate supplemented CPM matrices from sparse data, enabling the estimation of detailed CPM matrices based on historical data and correlations between electrical signals and catheter locations, reducing the need for extensive data collection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If comprehensive CPM matrices are generated using traditional data acquisition methods, then mapping precision is improved, but time consumption and labor intensity increase significantly

Engineering Contradiction:
Improvemapping precisionVSAvoidtime consumption
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary data collection and stores sparse CPM matrix data from multiple historical procedures in a database. This pre-collected data is then used to train a neural network model, eliminating the need to perform comprehensive data acquisition during each new procedure, thus reducing time consumption while maintaining mapping precision.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The neural network model learns from historical CPM matrix patterns and creates a computational copy of the mapping relationship. This model can then generate supplemented CPM matrices for new procedures by processing sparse input data, replacing the need for time-consuming direct measurements while preserving mapping accuracy.

Inventive Principle:
Principle #26Copying

2Measurement precision

If comprehensive CPM matrices are generated using traditional data acquisition methods, then mapping precision is improved, but procedural complexity increases

Engineering Contradiction:
Improvemapping precisionVSAvoidprocedural complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system replaces the mechanical data acquisition process (physical movement of catheter to multiple locations, manual measurement, and data compilation) with an automated neural network-based computational system. The neural network automatically processes sparse input data and generates comprehensive CPM matrices, significantly simplifying the procedure while maintaining or improving mapping precision.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Loss of time

If sparse CPM matrices are used directly, then time consumption is reduced, but mapping precision deteriorates

Engineering Contradiction:
Improvetime consumptionVSAvoidmapping precision
Core Design Contradiction:
Loss of timeVSMeasurement precision

Solution Approach 1:

The neural network transforms the parameter density of the CPM matrix by taking sparse input data (low parameter density) and generating supplemented output data (high parameter density). This parameter transformation allows the system to work with time-efficient sparse measurements while producing comprehensive high-precision mapping results through the learned relationships in the neural network.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4113528A1System and method to determine the location of a catheter
Publication Date: 2023.01.04 BIOSENSE WEBSTER (ISRAEL) LTD
  • EP4113528A1 patent drawingFigure 1
  • EP4113528A1 patent drawingFigure 2
  • EP4113528A1 patent drawingFigure 3

AI summary

Systems, devices, and techniques are disclosed for automatically generating CPM matrices. The system includes a processor configured to receive a plurality of historical, sparse CPM matrices and a plurality of historical, supplemented CPM matrices, wherein each sparse CPM matrix is associated with a respective supplemented CPM matrix; train a learning system based on the plurality of historical, sparse CPM matrices and the plurality of historical, supplemented CPM matrices, wherein the learning system is trained so as to generate a supplemented CPM matrix given a sparse CPM matrix; receive, by the trained learning system, a new, sparse CPM matrix; and generate, with the trained learning system, a new supplemented CPM matrix.