Download

Abstract

In this patent, we disclose a revolutionary system designed to enhance the accuracy and reliability of biosignal recordings, specifically electroencephalography (EEG) signals, though the principles are applicable to various EXG signal modalities. The hardware aspect of our invention centers around a novel apparatus tailored for sensing biosignals from a subject’s head. This apparatus consists of a net structure designed to sit comfortably on the subject’s head, incorporating a multitude of electrode structures interconnected by elastic legs. Each electrode structure houses a pair of electrodes: the first electrode faces towards the subject’s head to sense biosignals, while the second electrode is oriented away from the head, sensing environmental electrical signals. This dual-electrode design is crucial for mitigating noise from external sources. The electrodes are connected through elastic conduction elements integrated within the legs, which are insulated to prevent direct electrical contact with the user’s head. The net’s geodesic arrangement—where some electrodes are connected by multiple legs—ensures that the second electrodes are electrically interlinked, mirroring the electrical properties of the subject’s head, such as resistance and conductivity, thus further enhancing signal fidelity.

On the software and signal processing front, each electrode structure is equipped with an analog-to-digital converter and a multiplexer. These components digitize the biosignals and environmental electrical signals, respectively, before they are processed to extract clean EEG data. Our digital circuitry distinguishes the unique components of biosignals from noise by comparing the digitized signals from the first and second electrodes. Techniques such as Fourier transformation, independent component analysis, and other signal separation methods are employed to isolate and remove noise components, leaving behind the pure biosignal. Additionally, we incorporate a method that integrates signals from extra electrodes sensing muscular activity, which, when combined with environmental noise signals, forms a comprehensive noise profile. This profile is then refined to extract clean EEG data. Our innovative processing strategies, including real-time noise-rejection and immersive noise layer creation, significantly enhance the quality of EEG and EXG data. This not only improves the user experience by reducing setup time and increasing comfort but also opens new avenues for EEG application outside traditional laboratory settings, paving the way for advancements in brain-computer interfacing and beyond.

Dual-layer electrode structure for biosignal detection and noise cancellation

Analog to Digital Conversion Circuit for Biosignal Detection at each Electrode

Reverse Artifact Subspace Reconstruction for Noise Cancellation

Citation
@MISC{Shirazi2023-fz,
  title    = "System and methods for biosignal detection and active noise
              cancellation",
  author   = "Shirazi, Seyed Yahya",
  journal  = "US Patent",
  number   = "20230240581:A1",
  month    =  aug,
  year     =  2023
}