Biomagnetic Interference Suppression via Signal Space Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods are inadequate for efficiently eliminating interference signals from biomagnetic measurements when the source is located close to the object being measured, such as near the head or neck, as they cannot clearly classify these sources as external or internal, making it difficult to suppress interferences using existing shielding or signal processing techniques.
Innovation Solution
A method that utilizes Signal Space Separation (SSS) reconstruction to identify and separate internal and external signals, combined with statistical analysis and component decomposition techniques like PCA, ICA, and SVD, to eliminate interference signals from the intermediate space near the sensors, using a system with feedback coils and reference sensors to isolate and filter out interference sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the object being measured is placed as close as possible to the sensors to improve signal-to-noise ratio, then measurement sensitivity is improved, but interference from sources near the object (such as patient's head or neck) increases
Solution Approach 1:
The patent divides the measurement space into distinct regions: volume V1 containing the object being measured, volume V2 containing the sensors, and volume V3 containing external interference sources. This spatial segmentation allows different signal processing strategies to be applied to signals originating from different regions, enabling the system to maintain close sensor-object placement while selectively suppressing interference from specific directions and regions.
Solution Approach 2:
The patent introduces an intermediate processing layer between the raw sensor signals and the final measurement output. This layer performs SSS reconstruction and signal space separation, acting as a mediator that distinguishes between signals from the object, nearby interference sources, and external interference. This intermediary processing enables the system to maintain high sensitivity while filtering out harmful interference through mathematical decomposition of the measurement space.
2Object-affected harmful factors
If passive magnetic shielding rooms are used to suppress external magnetic fields, then shielding effectiveness is improved, but device complexity and measurement environment requirements increase
Solution Approach 1:
The patent replaces the mechanical/passive shielding approach with a computational/electronic solution. Instead of relying on physical shielding rooms with complex magnetic shielding structures, the system uses SSS reconstruction and signal space separation algorithms to mathematically separate and suppress interference signals. This substitution of mechanical shielding with computational processing reduces device complexity and eliminates the need for specialized measurement environments while maintaining effective interference suppression.
3Object-affected harmful factors
If gradiometers are used to achieve magnetic shielding against distant sources, then shielding factor is improved, but ability to suppress nearby interference sources deteriorates
Solution Approach 1:
The patent employs dynamic signal processing that adapts to the specific characteristics of interference sources in real-time. Unlike fixed gradiometer configurations that are optimized for distant sources, the SSS reconstruction method dynamically separates signals based on their spatial and temporal characteristics, allowing the system to effectively suppress nearby interference sources while maintaining sensitivity to biomagnetic signals. The method adjusts its interference suppression strategy based on the actual measurement conditions rather than relying on fixed geometric configurations.
4Object-affected harmful factors
If SSS method is used to separate biomagnetic signals from external interferences, then external interference suppression is improved, but ability to handle nearby interference sources (intermediate space) remains insufficient
Solution Approach 1:
The patent applies different signal processing strategies to different spatial regions: signals from volume V1 (object), volume V2 (sensors), and volume V3 (external sources) are handled with region-specific approaches. For the intermediate region near the sensors, the patent uses enhanced SSS reconstruction combined with statistical analysis and component decomposition techniques that are specifically tailored to handle the unique characteristics of nearby interference sources, providing locally optimized interference suppression for each spatial zone.
Solution Approach 2:
The patent combines multiple signal processing methods into a composite approach: SSS reconstruction is integrated with statistical analysis, principal component analysis (PCA), independent component analysis (ICA), and singular value decomposition (SVD). This composite methodology leverages the strengths of each individual technique to comprehensively handle interference from all regions, including the challenging intermediate space near the sensors, achieving superior interference suppression that no single method could accomplish alone.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Effectively eliminates interference signals from the intermediate space, improving the signal-to-noise ratio in biomagnetic measurements by accurately distinguishing and removing signals from sources near the measurement sensors, enhancing the accuracy of biomagnetic signal detection.
Implementation Method 1
a magnetic field measured by a multi-channel MEG device is analysed by examining three different volumes of the measurement geometry
Implementation Method 2
the interference is measured in the vicinity of the region being shielded by means of a sensor or sensors; and based on this measurement, the interference field is compensated with current-carrying coils that produce a magnetic field
Data Source
Figure 1
Figure 2
AI summary
The present invention recognises and eliminates from a biomagnetic measurement signal interferences whose source is disposed in the direct vicinity of an object being measured. The invention utilises the SSS method that can be used to separate from one another the signals associated with the internal and external sources of a set of measurement sensors by calculating two series developments. The sources to be examined in the invention and disposed in the so-called intermediate space produce a component to both of the developments, and can, therefore, be detected by means of an analysis to be performed in a time domain. This division into components can be made using the Principal Component Analysis (PCA), the Independent Component Analysis (ICA) or the Singular Value Decomposition. Finally, the clarified interferences in the intermediate space can be eliminated from the measured signal using, for example, the linear algebraic orthogonal projection.