Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GAPses: Versatile smart glasses for comfortable and fully-dry acquisition and parallel ultra-low-power processing of EEG and EOG

Sebastian Frey, Mattia Alberto Lucchini, Victor Kartsch, Thorir Mar Ingolfsson, Andrea Helga Bernardi, Michael Segessenmann, Jakub Osieleniec, Simone Benatti, Luca Benini, Andrea Cossettini

TL;DR

GAPses introduces a versatile smart-glasses platform that enables fully dry, comfortable EEG and EOG acquisition with direct electrode-electronics interfacing and onboard edge processing via a GAP9 SoC. It demonstrates multi-channel EEG (8) and EOG (3) sensing, high-performance on-device inference (down to tens of microjoules per inference), and two strong use cases: EOG-based eye-movement classification and EEG-based BrainMetrics biometrics, achieving state-of-the-art-like accuracy with low power. The work validates across alpha, SSVEP, and motor-movement EEG paradigms and shows robust eye-movement classification even under motion; it also compares favorably to existing ExG glasses and provides open-source designs to accelerate adoption. Overall, GAPses advances wearable neurotechnology by combining unobtrusive form factor, dry-electrode comfort, and energy-efficient edge computation for real-time brain–computer interfacing.

Abstract

Recent advancements in head-mounted wearable technology are revolutionizing the field of biopotential measurement, but the integration of these technologies into practical, user-friendly devices remains challenging due to issues with design intrusiveness, comfort, and data privacy. To address these challenges, this paper presents GAPses, a novel smart glasses platform designed for unobtrusive, comfortable, and secure acquisition and processing of electroencephalography (EEG) and electrooculography (EOG) signals. We introduce a direct electrode-electronics interface with custom dry soft electrodes to enhance comfort for long wear. An integrated parallel ultra-low-power RISC-V processor (GAP9, Greenwaves Technologies) processes data at the edge, thereby eliminating the need for continuous data streaming through a wireless link, enhancing privacy, and increasing system reliability in adverse channel conditions. We demonstrate the broad applicability of the designed prototype through validation in a number of EEG-based interaction tasks, including alpha waves, steady-state visual evoked potential analysis, and motor movement classification. Furthermore, we demonstrate an EEG-based biometric subject recognition task, where we reach a sensitivity and specificity of 98.87% and 99.86% respectively, with only 8 EEG channels and an energy consumption per inference on the edge as low as 121 $μJ$. Moreover, in an EOG-based eye movement classification task, we reach an accuracy of 96.68% on 11 classes, resulting in an information transfer rate of 94.78 bit/min, which can be further increased to 161.43 bit/min by reducing the accuracy to 81.43%. The deployed implementation has an energy consumption of 40 $μJ$ per inference and a total system power of only 12.4 mW, of which only 1.61% is used for classification, allowing for continuous operation of more than 22 h with a small 75 mAh battery.

GAPses: Versatile smart glasses for comfortable and fully-dry acquisition and parallel ultra-low-power processing of EEG and EOG

TL;DR

GAPses introduces a versatile smart-glasses platform that enables fully dry, comfortable EEG and EOG acquisition with direct electrode-electronics interfacing and onboard edge processing via a GAP9 SoC. It demonstrates multi-channel EEG (8) and EOG (3) sensing, high-performance on-device inference (down to tens of microjoules per inference), and two strong use cases: EOG-based eye-movement classification and EEG-based BrainMetrics biometrics, achieving state-of-the-art-like accuracy with low power. The work validates across alpha, SSVEP, and motor-movement EEG paradigms and shows robust eye-movement classification even under motion; it also compares favorably to existing ExG glasses and provides open-source designs to accelerate adoption. Overall, GAPses advances wearable neurotechnology by combining unobtrusive form factor, dry-electrode comfort, and energy-efficient edge computation for real-time brain–computer interfacing.

Abstract

Recent advancements in head-mounted wearable technology are revolutionizing the field of biopotential measurement, but the integration of these technologies into practical, user-friendly devices remains challenging due to issues with design intrusiveness, comfort, and data privacy. To address these challenges, this paper presents GAPses, a novel smart glasses platform designed for unobtrusive, comfortable, and secure acquisition and processing of electroencephalography (EEG) and electrooculography (EOG) signals. We introduce a direct electrode-electronics interface with custom dry soft electrodes to enhance comfort for long wear. An integrated parallel ultra-low-power RISC-V processor (GAP9, Greenwaves Technologies) processes data at the edge, thereby eliminating the need for continuous data streaming through a wireless link, enhancing privacy, and increasing system reliability in adverse channel conditions. We demonstrate the broad applicability of the designed prototype through validation in a number of EEG-based interaction tasks, including alpha waves, steady-state visual evoked potential analysis, and motor movement classification. Furthermore, we demonstrate an EEG-based biometric subject recognition task, where we reach a sensitivity and specificity of 98.87% and 99.86% respectively, with only 8 EEG channels and an energy consumption per inference on the edge as low as 121 . Moreover, in an EOG-based eye movement classification task, we reach an accuracy of 96.68% on 11 classes, resulting in an information transfer rate of 94.78 bit/min, which can be further increased to 161.43 bit/min by reducing the accuracy to 81.43%. The deployed implementation has an energy consumption of 40 per inference and a total system power of only 12.4 mW, of which only 1.61% is used for classification, allowing for continuous operation of more than 22 h with a small 75 mAh battery.
Paper Structure (37 sections, 3 equations, 6 figures, 6 tables)

This paper contains 37 sections, 3 equations, 6 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Biopotential data acquisition platforms
  4. Smart glasses for EEG and EOG sensing
  5. System Design
  6. Electrodes: design and placement
  7. EOG electrodes
  8. EEG electrodes
  9. Electrodes-frames interface
  10. Glasses frame
  11. Electrodes interface PCB
  12. BioGAP data acquisition platform and channel selection interface
  13. Baseboard
  14. Bio-potential expansion board
  15. Channel selection interface
  16. ...and 22 more sections

Figures (6)

  • Figure 1: (A) Photo of the whole system. (B) Detailed view of the custom EOG (flat) and EEG (brush) electrodes. (C) Electrode interface PCB, embedded in the glasses' frame. (D) Channel-selection interface PCB, used to select which channels (among the 3 EOG and 8 EEG channels available) are interfaced to BioGAP. Two alternative versions are designed, resulting in an EEG-only configuration or hybrid EEG+EOG configuration. (E) BioGAP acquisition and processing platform. The bottom block diagram shows a simplified circuit diagram and the connectivity between the different PCBs and their internal key components. Bottom right: photo of a subject wearing GAPses
  • Figure 2: Measurement of the horizontal and vertical EOG signal while different eye movements are performed.
  • Figure 3: Spectrogram of the EEG system evaluation by performing alpha waves measurement in the eyes open vs eyes closed experiment.
  • Figure 4: (A) NCCA of the SSVEP experiment showing the response at different window lengths (average values across five subjects). (B) corresponding CCA for one subject.
  • Figure 5: Averaged Accuracy and ITR values for the subject-specific models when considering different fractions of the 2-second windows. Inset: t-SNE visualization of the highest accuracy point.
  • ...and 1 more figures