Eye-writing communication for patients with amyotrophic lateral sclerosis

Jang Zern Tsai, Tsai Shih Chen

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

The eye-writing method predefines a symbol set containing symbols with distinct writing traces. A user of this method rotates his or her eye balls to "write" a symbol according to its designated writing trace. Meanwhile, the eye movement is detected using a suitable technique such as the electro-oculography, which measures voltage differences on the skin around the user's eyes. Distinct features of the acquired eye-movement signals are extracted in order to determine which symbol, among those in the symbol set, the user's eyes have just written. An eye-writing system has been implemented in this study. Tests on subjects with no known disabilities have been conducted and the performance has been evaluated. The study found that eye-writing system is potentially useful for facilitating communication of sever ALS patients who have lost most of their oral speaking and handwriting abilities.

Original languageEnglish
Title of host publicationASSETS'09 - Proceedings of the 11th International ACM SIGACCESS Conference on Computers and Accessibility
Pages239-240
Number of pages2
DOIs
StatePublished - 2009
Event11th International ACM SIGACCESS Conference on Computers and Accessibility, ASSETS'09 - Pittsburgh, PA, United States
Duration: 26 Oct 200928 Oct 2009

Publication series

NameASSETS'09 - Proceedings of the 11th International ACM SIGACCESS Conference on Computers and Accessibility

Conference

Conference11th International ACM SIGACCESS Conference on Computers and Accessibility, ASSETS'09
Country/TerritoryUnited States
CityPittsburgh, PA
Period26/10/0928/10/09

Keywords

  • ALS
  • Communication
  • Electro-oculography
  • Eye
  • Writing

Fingerprint

Dive into the research topics of 'Eye-writing communication for patients with amyotrophic lateral sclerosis'. Together they form a unique fingerprint.

Cite this