June 2020
Volume 61, Issue 7
ARVO Annual Meeting Abstract  |   June 2020
Visual lifestyle of myopic children assessed with AI-powered wearable monitoring
Author Affiliations & Notes
  • Michael Mrochen
    Vivior AG, Zurich, Switzerland
  • Pavel Zakharov
    Vivior AG, Zurich, Switzerland
  • Burcu Nurözler Tabakcι
    Faculty of Medicine, Biruni University Hospital, Istanbul, Turkey
  • Cafer Tanrιverdi
    Faculty of Medicine, Medipol University, Istanbul, Turkey
  • Aylin Kιlιç
    Faculty of Medicine, Medipol University, Istanbul, Turkey
  • Daniel Ian Flitcroft
    Children’s University Hospital, Dublin, Ireland
  • Footnotes
    Commercial Relationships   Michael Mrochen, VIVIOR AG (I); Pavel Zakharov, VIVIOR AG (E); Burcu Tabakcι, None; Cafer Tanrιverdi, None; Aylin Kιlιç, VIVIOR AG (C); Daniel Flitcroft, VIVIOR AG (C)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2020, Vol.61, 82. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Michael Mrochen, Pavel Zakharov, Burcu Nurözler Tabakcι, Cafer Tanrιverdi, Aylin Kιlιç, Daniel Ian Flitcroft; Visual lifestyle of myopic children assessed with AI-powered wearable monitoring. Invest. Ophthalmol. Vis. Sci. 2020;61(7):82.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Purpose : To investigate visual behaviour of myopic children (6-16 years old) using a novel multi-sensor wearable monitor

Methods : The visual behavioural profiles of 54 myopic children (34 girls and 20 boys, mean refraction of -3.7D in spherical equivalent) were collected with the Vivior Monitor. The profiles represent the occurrence of the different distances to the objects of visual activity of the user (vergence distance), in the range from 0.2 to 2 meters, as well as illumination conditions, motion patterns, head orientations. The automatic recognition of the activities performed by subject was performed using machine learning algorithms, which provided the identification of the type of visual activity, such as handheld media, desktop work, computer work, viewing, etc. We have evaluated the usage patterns of the Monitors and compared the measurements based on degree of myopia, gender and age.

Results : The Vivior Monitor was well tolerated. An average of 37.8 hours of data was obtained from subjects (min-max: 8.7 – 93.7). We observed a large contribution of extreme near vision zone (below 25 cm), 20% of total time. Significant variations in behaviour were observed in younger (<=11 yrs) and older (>=12 years) children. The amount of physical activity is reduced by 13% in the older group (p = 0.005), far zone (above 1 meter) usage reduced 16% in the older group (p = 0.01) and time spent on the computer increased more than 2 times (p = 0.045). The only observed significant difference in gender was the time spent on computer: 0.76 hours per day in average for boys versus 0.29 hours per day for girls (p = 0.045). No illumination related differences have been statistically significant in any of the analysed splits.

Conclusions : The Vivior Monitor is capable of providing insights into the behavioural patterns of children. It allows investigation of lifestyle from the different perspectives, such as visual distances, illumination conditions, position and activities. Statistically significant differences were observed with age and gender, with old children spending less time at distant viewing, more computer use and less physical movement. Boys spent significantly more time on computer tasks than girls. Future studies will examine the predictive value of such behavioural data for myopic onset and progression.

This is a 2020 ARVO Annual Meeting abstract.


This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.