Yifan Xu

Yifan Xu

he/him

Undergraduate student interested in designing optimization algorithms and writing high-performance implementations to solve large-scale linear and nonlinear programming problems.

Evaluation of SSVEP Stimuli Design for Visual Field Assessment

Status: Published — 2022 International Conference on Cyberworlds (CW) · Co-author: Aung Aung Phyo Wai · Affiliation: National University of Singapore High School of Mathematics and Science & Nanyang Technological University

Steady-State Visual Evoked Potential (SSVEP) brain–computer interfaces can leverage frequency-tagged responses from the visual cortex not only for communication but also for vision assessment — including early detection of visual-field loss from diseases such as glaucoma. However, stimulus designs optimized for visual spellers differ substantially from what visual-field assessment requires, and prior to this work no study had evaluated SSVEP stimulus designs specifically for discriminating abnormal from normal visual fields.

We propose three stimulus-layout designs and evaluate them across normal and simulated-abnormal visual-field scenarios with twenty-one healthy subjects, assessing both usability and classification performance (using Canonical Correlation Analysis features). The multifocal layout achieved the highest mean accuracy (86.88 ± 1.47%) at discriminating normal from two abnormal visual fields, while subjects’ qualitative preferences — driven largely by visual comfort — favored the concentric design. We conclude that the SSVEP multifocal layout can detect changes in visual-field characteristics while supporting comfortable, objective visual-field assessment.

Materials

  • Paper — Evaluation of Steady-State Visual Evoked Potentials (SSVEP) Stimuli Design for Visual Field Assessment, 2022 International Conference on Cyberworlds (CW): IEEE Xplore (DOI)