Hacking My Vision with Amblyopia Apps


A Current Biology correspondence (Li et al, 2013 Current Biology Vol 23 No 8 R308) recently discussed a novel treatment for ambylopia, a vision problem which a quick Google search shows affects up 1-5% of the US population, and yours truly in the past!

Amblyopia, also known as "lazy eye" is a condition in which the brain begins to heavily favor the input of one eye over the other. Left untreated, this can result in blindness in the weak eye. Historically, the condition has been corrected by patching the strong eye, forcing the patient to use the input from the bad eye exclusively. Over time, this teaches the brain to use signals from the other eye, but it's a slow and frustrating process.

I was treated this way--- patching an eye--- when I was growing up, and I don't remember much other than absolutely hating it! In the paper patching is called 'monocular learning' --- and this traditional treatment was shown to be 4x less effective than the new one they used.

The paper employed a method called 'dichoptic learning' which involves sending separate parts of one image to each eye, and then performing tasks which require knowledge of the full image. This sounds hard to replicate in practice, but there is a pretty simple at-home way to try it---just grab a pair of ordinary red-blue 3D glasses, and then try to perform tasks with red and blue objects. The red and blue lenses make it very hard for each eye to see objects that are red and blue, respectively---especially in low-contrast environments. In the study, I believe they played tetris, but in theory you could play any game, or perform any task, really, so long as you were using both eyes together.

Well, before jumping into the research further it makes sense to learn how to measure success. So how could they tell the patients were getting better? Do you remember those eye charts where the doctor asked you to read back letters? Well, each line has a score, and your point total is the score for the best line you read, plus 0.1, minus 0.02 times the number of letters you got on that line. It's called your logMAR score, and is used to measure visual acuity. In the study, the folks who played the tetris games on average went from 0.5 logMAR to 0.3 over just two weeks of playing an hour a day, corresponding to an improvement from 20/60 vision to 20/40! They massively improved their vision with minimal effort!

In a recent visit to my opthamologist, my doctor mentioned I might be at risk for returning lazy eye as I age. A call home to mom confirmed that this had also happened to a grandparent in my family, so some preventative action is warranted. And when I had this conversation, there happened to be a few news articles about the study above and vision therapy floating around the internet.

So, I snagged some 3D glasses and DIY'd a few apps together to mess around with it.

My first move was forking a javascript tetris game and coloring the blocks red and blue. Pro tip---use the sliders at the top of the page to set the background color to be max red, max blue, no green (#FF00FF) for minimum contrast and a hard game.

I played tetris for a couple hours, and anecdotally, my vision seemed much sharper the day after than I was used to. So I put a little more effort into it and wrote a couple more things.

I wanted to be able to train my brain while I browsed the internet, so I made a Chrome extension that could color text on internet pages randomly red and blue as well. install the extension, click the button, and wham---your internet forum now trains your brain as well. Or at least that was the idea---in practice, the extension is a little visually unpredictable. Some pages work nicely, others, not so much.

After using the chrome extension for a bit, I got sick of the unpredictable visual results for random pages, so I created a simple page with consistent look for the vision task quote-reading app.

For my next move, I'm going to grab some eye charts and start tracking my logMAR score improvement. The effects in the study seemed significant enough, with fairly minimal effort required, that I should be able to measure an improvement at home.

The next time I visit my opthamologist, I hope to have an opinion about the amblyopia tetris research, some data and an interesting story.