rss
Br J Ophthalmol doi:10.1136/bjo.2010.198135
  • Innovations

A novel paediatric game-based visual-fields assessor

  1. Peng T Khaw1
  1. 1National Institute for Health Research Biomedical Research Centre, Moorfields Eye Hospital and University College London Institute of Ophthalmology, London, UK
  2. 2National Institute for Health Research Biomedical Research Centre, Manchester Royal Eye Hospital and University of Manchester, Manchester, UK
  1. Correspondence to Mr Tariq M Aslam, National Institute for Health Research Biomedical Research Centre, Manchester Royal Eye Hospital and University of Manchester, UK; tariq.aslam{at}cmft.nhs.uk
  • Accepted 24 February 2011
  • Published Online First 3 April 2011

Abstract

Aim To determine the feasibility of using a computer game to measure visual fields in children.

Methods The authors describe the development and assessment of a novel computer-game apparatus to measure visual fields in children. It is based upon a computer game visible on a monitor, housed in an interactive model-castle structure.

Results The authors conducted a total of 25 field tests with the final apparatus, in 25 eyes of 19 children, aged 4–14 years. On two occasions children failed to complete the exam, owing to hardware and software defects that were subsequently rectified. For the 23 completed fields, the median time for completion of the full test was 4.5 min. 16 out of 18 clinically normal patients showed normal fields; the only failures were in two 4-year-old children who still managed to complete fields, but with generalised reduced responses that were deemed abnormal by our predetermined criteria. For the five eyes with expected glaucomatous loss, all five visual fields measured were completed and showed abnormal fields consistent with their medical condition. Positive feedback was given from all children about their testing experience.

Conclusions It is feasible to develop a computer-game-based system to measure fields in children in a non-invasive, affordable and entertaining way.

Measurement of visual fields can be prohibitively difficult in children, particularly below the age of 10–11 years.1 2 Problems include lack of central fixation,3 short attention span3–5 and intolerance. Recent systems have demonstrated user interaction2 6 may improve tolerance and reliability in children and have demonstrated the potential for personal computer testing.7 Other recent innovations include eye tracking technology, proposed in 19898 and developed at the University of Tokyo.9 However, this requires expensive apparatus and has not yet been validated.10

Our aim was to determine whether a computer-game system could be devised to measure visual fields in children while they play and enjoy an entertaining game.

Methods

The study had full ethics committee approval, and informed consent was obtained from all participants and parents. The development of the game-based visual-fields assessor relied on multidisciplinary input and combined international perimetric society standards11 with computer game science. The final system uses a supra-threshold examination strategy with fixed intensity. Background luminance and other psychophysical constants were set for this feasibility study based on literature review and estimations from adult trials of the test.

The software was developed using BBC BASIC/Windows v5.80a and was conceived and written by the first author. Game design acknowledged Bateman and Boon's12 game play types, and Salen and Zimmerman13 and Mulligan and Patrovsky14 theories on player psychologies. We adopted the iterative game-design process advocated by Salen and Zimmerman13 and Fullerton et al,15 which relies on gaining feedback from players early on and repeated game redesign—our game resulted from 17 cycles of appraisal and reprogramming. Development included designing instructions for explaining how to play the ‘Crazy Castle’ game and a cohesive storyline that ties in all that is required of the player.

The complete game structure consists of a large wooden castle (1 m3) with a set of occludable spectacle frames attached to a large central window on an exterior castle wall (figure 1). A closed wooden drawbridge barrier obstructs view through the window into the castle. When a child's forehead is correctly aligned in the glasses (figures 2 and 3) a switch activates opening of the drawbridge which then reveals a computer game on a screen (figure 3) inside the castle, at a set distance from the spectacle frames. If the correct alignment is lost, the drawbridge instantly closes, hiding the game screen from view.

Figure 1

Paediatric game-based vision assessor.

Figure 2

Child waiting to play game, holding control pad then correctly positioned to see and play game.

Figure 3

Examples of images seen in game: (A) tomato correctly positioned to mash; (B) tomato being mashed; (C) peripheral ghost being lasered.

The child uses an attached keypad to play the game, which is separated into four stages, allowing for the child to learn to play while maintaining position and for the software to set individualised variables. The game premise involves a small central character of a wizard with a plunger. When a tomato hovers under this plunger, the child presses the appropriately textured console button to squash the tomato (central fixation game). This central game is moved to different corners of the screen, and peripheral field targets appear intermittently as ghosts in remaining corners. The child must continue the central fixation game but is encouraged to press a separate button when they see these peripheral ghosts (peripheral target game). These ghosts (figure 4) appear at 10, 20 and 30° from the central fixation game at each of four axes (see figures 4 and 5). They are presented for 200 ms and increase in size if no response is provoked. Each tested peripheral target is graded 0 (not seen), 1 (5 mm size seen), 2 (4 mm), 3 (3 mm) or 4 (2 mm).16 Missed stimuli are responded to with repeat testing. In accordance with perimetric standards and computer-game theory, there are multiple programming measures, animations, sound effects and graphics incorporated to optimise central fixation importance, reduce false negatives and false positives, and optimise utility of the game. The examiner is able to monitor progress on a separate laptop adjacent to the castle. Once tested, the children were invited to respond to a questionnaire about their experience with the visual-field exam.

Figure 4

Discrimination potential. Fields measured of 14-year-old boy with no ocular pathology in the right eye, but with glaucoma and suspected field defects found in his left eye that corresponded with those found on Humphrey testing. Note that in the right eye, almost all targets were seen normally (score 3 or 4). In the left eye, there are several areas of lost field (seven out of the 24 tested areas score less than 3). The left eye field was measured after the right eye.

Figure 5

Acceptability of test. This 11-year-old child had documented poor compliance with standard fields tests, and none were found acceptable. Goldman fields were not successful, and Humphreys fields showed multiple areas of field loss with poor fixation. On his first attempt with our game-based fields analyser, however, he returned good peripheral fields while demonstrating good central target fixation, in accordance with his clinically expected visual function.

Results

Nineteen children were recruited from an outpatient department at the Richard Desmond Children's Eye Centre at Moorfields Eye Hospital. Tests were carried out in the Fight for Sight NIHR Children's Research Facility. Two children failed to complete the exam owing to hardware and software failures that were subsequently corrected.

Twenty-three fields were completed, in 23 eyes of 17 children (six bilateral). All completed fields had physiological blind spots found. Normal fields were defined through earlier pilot studies on normal adults. The median time for completion of the tests including training stages was 4.5 min (range 3.11–8.0 min). There were eight girls and nine boys, aged 4–14 years (five children aged 4–6 years, eight aged 7–10 years and four aged 10–14 years).

Two 4-year-olds with expected normal fields showed a general delay in response which, according to our preset definitions, was classed as abnormal. The problems for these two children included, first, requiring a greater time to learn the game than was currently programmed into the system. Furthermore, the requirement of two separate buttons to respond in the course of the game was perhaps too complex, and a single button option might be more apt in this age group. However, it was very promising that they both still fully completed the tests, and we feel that reducing game difficulty for this age would counter the problems that arose.

For all 15 other eyes with clinically expected normal fields (all normal eyes of children aged over 4 years), the children were, at their first attempt, able to complete the assessment giving normal visual fields. This included three children aged 6 years and under. For the five eyes with expected glaucomatous loss, all five visual fields measured showed abnormal fields. Discrimination potential and acceptability of test are demonstrated in figures 4 and 5. In questionnaires, all children enjoyed playing the game and gave positive feedback.

Discussion

The gaming system has shown great promise, in this feasibility study, to provide clinically useful information in a way that is accessible and enjoyable for the child with no discomfort or potential for harm. Developments need to be made in terms of shortfalls in hardware and software as well as in optimising the psychophysical consistency and validity of peripheral targets. Ultimately, we anticipate this system being particularly useful for perimetry in targeted groups of children in terms of age, aptitude, personality and physical ability. Formal validation and further reliability studies will be required before the system can be presented as a fully validated and reliable clinical tool. The final technology would potentially have a greater diagnostic precision and sensitivity to identify visual-fields loss than existing tests. The test would be quicker and more reliable, and therefore be a lesser drain on resources. It would be more affordable than alternative proposals in development and more enjoyable to experience for the child than any current system. It has potential for worldwide adoption owing to its basis upon widely available and affordable technology. In modified form, the game-based system could also be utilised for other visual-function measurements such as acuity and contrast sensitivity.

Acknowledgments

We would like to acknowledge input and comments from professionals including optometrists, ophthalmologists, vision scientists and paediatric nurses over the course of this study. Young members of the public also gave key input during the development phase of the game, with particularly useful and lengthy input from H Jessa and Z Jessa.

Footnotes

  • Funding National Institute for Health Research Biomedical Research Centre Moorfields Eye Hospital and University College London Institute of Ophthalmology and Moorfields Trustees.

  • Competing interests None.

  • Patient consent Obtained from the parents.

  • Ethics approval Ethics approval was provided by the Moorfields Eye Hospital.

  • Provenance and peer review Not commissioned; externally peer reviewed.

References

Relevant Article

Register for free content


Free sample
This recent issue is free to all users to allow everyone the opportunity to see the full scope and typical content of BJO.
View free sample issue >>

Don't forget to sign up for content alerts so you keep up to date with all the articles as they are published.