Synthetic-Data MCG Denoising in Unshielded Magnetic Environments

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing denoising techniques for magnetocardiography (MCG) in unshielded environments are inadequate due to complex and non-stationary magnetic noise sources, leading to reduced measurement accuracy and instability.

Innovation Solution

A deep learning network trained on synthetic magnetocardiography data is used to separate magnetic signals from noise, utilizing inertial measurement units (IMUs) to track magnetometer motions and generate training data for effective denoising.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If magnetocardiography is performed in magnetically unshielded environments, then ease of operation and clinical applicability are improved, but measurement precision deteriorates due to complex and non-stationary magnetic noise sources

Engineering Contradiction:

Improveclinical applicabilityVSAvoidmeasurement accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent introduces an intermediary deep learning network that acts as a mediator between the noisy magnetic measurements from unshielded environments and the clean signals needed for accurate diagnosis. The network processes and filters the complex non-stationary noise, enabling clinical use without magnetic shielding while preserving measurement accuracy.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical/physical approach of magnetic shielding with a computational approach using deep learning. Instead of using physical barriers to block magnetic noise, the system uses synthetic data training and neural networks to identify and remove noise components, achieving both portability and precision.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Device complexity

If conventional denoising techniques are used, then device complexity is reduced, but reliability deteriorates due to inadequate handling of complex and non-stationary noise sources

Engineering Contradiction:

Improvedenoising algorithm simplicityVSAvoiddenoising effectiveness

Core Design Contradiction:

Device complexityVSReliability

Solution Approach 1:

The patent applies preliminary action by training the deep learning network on synthetic magnetocardiography data before actual clinical use. This pre-training phase allows the network to learn various noise patterns and characteristics, making it more reliable when processing real measurements from unshielded environments without requiring complex adaptive algorithms during operation.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20250375136A1Removing magnetocardiography noise using a deep learning network trained on synthetic magnetocardiography

Publication Date: 2025.12.11 SB TECH INC

AI summary

The present disclosure relates to methods, systems, and apparatus, including computer programs encoded on computer storage media, for denoising magnetic measurements. An example method includes obtaining, using a plurality of magnetometers of a magnetically unshielded device, noisy magnetic measurement data of a magnetic field at least partially caused by an organ of a subject; obtaining, using one or more inertial measurement units (IMUs) of the magnetically unshielded device, a motional measurement of the plurality of magnetometers; determining cleaned magnetic field data based at least in part on the noisy magnetic measurement data and the motional measurement data; and taking an action based at least in part on the cleaned magnetic field data.