Louise Allen, Addenbrookes
Contact: "Allen, Louise" <louise.allen@addenbrookes.nhs.uk>
2016 agreed
Louise notes: If the test is to be useful in clinical practice the matching process is important a 50/50 chance of getting the correct answer won't do!
earlier suggestion
DIY eye-tests
Many people would benefit from a way to take calibrated eye tests at home or in public places, rather than requiring special equipment and professional supervision. Your task is to design a visual acuity test that can be set up and administered using two screens - one four metres away, and one in the user's hand (their own phone or tablet). The distant screen could be anything capable of running a web browser, such as a TV or public display, with authenticated coordination to the user’s phone via a separate HTTP server. Once the test starts, users will need to operate the local device without looking at it - using easy touch controls, with instructions given by audio. You’ll need to investigate the relevant standards and procedures to ensure that the result is clinically valid.
Original background
There are two projects, the simplest is a DIY visual acuity testing system for hospitals / vision screening in the community
We see over 50,000 patients a year in the eye out patients dept at Addenbrooke's and this sort of number is typical for a large teaching hospital. Every patient has to have a test of visual acuity (threshold of resolution, tested in a clinic setting using a Snellen or LogMAR chart) before they have other tests and their consultation. This is the bottleneck of the clinic because a nurse currently has to do every visual acuity test and we only have 2 nurses to do this despite having 8 doctors seeing patients in clinic. Each test takes about 10 minute so you can see how easy it is to fall behind.
I estimate that about 30% of patients would be able to assess their own visual acuity if given an automated process, which is what I am looking to develop. My idea is to have a touch screen tablet in front of the patient which drives a second monitor 4 metres away. Optotype letters would be shown on the distant screen, decreasing in size using a staircase algorithm and the patient would match the letter to one of a number displayed on the tablet screen in front of them until they make 50% errors. This would then give a threshold visual acuity which would be recorded.
The optotype standards / sizes are well established. Auditory and visual instructions would need to inform the patients what to do. We could use accepted optotypes for kids - who would probably respond more positively to this as a game than most of the elderly.
I imagine that it might be a little like the automated check outs in supermarkets, gradually people will prefer to use the system so they get through quicker!
The second project is to work on a visual acuity testing system for
pre-verbal children. I have previously developed a system called
KidzEyez which is now licenced which uses a webcam feed from the centre
of a display screen to monitor the responsive eye movements of an
infant to cartoon targets appearing in the periphery of the screen -
which informs us about the peripheral visual field of the infant. I have
the prototype at Addenbrooke's. Hazel Kay (who developed the established
Kays pictures visual acuity cards for children) and I would like to
develop a visual acuity test along the same lines - giving the infant
two targets of diminishing size to distinguish between and measuring
visual acuity as the smallest size of optotype to which the child sees
and looks towards. See
https:www.tes.com/teaching-resource/kidzeyez-6380245 for info about
Kidzeyez.
