December 8th, 2014 | Published in eHealth & Assisted Living
Every year I get to sit through a large number of presentations from people who have developed new products or applications. Whilst lots of them are interesting, it‘s rare to see one that jumps out. This year I was fortunate to sit through one of those. It was given by Andrew Bastawrous from the London School of Tropical Medicine, describing a smartphone adaptor and application he and his colleagues had designed to help diagnose sight problems. It was a master class in how to make decisions about hardware and application development.
Andrew has now set up Peekvision – a company planning to make and distribute their product. They’ve started an IndieGoGo campaign to fund it, which will let them supply them to community workers around the world who have been selected by their partners Médecins sans Frontières and the International Agency for the Prevention of Blindness. For £60 you can pay for a device to be given to a healthcare worker. I can’t think of a better Christmas present to give.
What was so good about Andrew’s talk was the pragmatic approach to product design. It started by identifying a real problem that needed to be solved and then addressing the challenges of rolling it out in a way that made sense to patients, the local community and health workers. So many developers would have just reached for their iPhone and started to write code. This brilliantly demonstrates why that would have been the wrong approach.
Andrew’s presentations are well worth listening to. He’s done a short TED talk, but if you’re a developer, head to the one he gave at the Cambridge Wireless “Future of Wireless Conference” last June. It’s only audio, but it tells the story. The background is that they wanted to find a way to test sight in rural Kenya. They’d discovered that carrying around several tens of thousand pounds-worth of the high tech equipment found in most western hospitals didn’t work – they needed something that anyone could use, which was rugged and which didn’t scare the patient.
I’d assumed that in countries like Kenya the main challenge would have been lack of eye surgeons. Although there’s a far lower percentage than we expect in Europe or the US, the bigger problem is identifying patients and bringing them in for surgery, especially with a large rural community. Surgeons end up travelling to eye clinics, rather than concentrating on treating patients. So Andrew realised they needed something that would allow anyone to perform the testing, providing a regular supply of patients for the surgeons.
Their choice of phone and platform was pragmatic. With unreliable and unpredictable power and transport systems the first problem is battery life. The phone needs to work as soon as it arrives. That means a smartphone with a removable battery, so that a spare is always being charged from a small solar panel. It’s one of those small, important details that you need to get right at the earliest possible stage of your design process.
Peekvision’s aim is that the phone can be used by anyone. For eye tests it lets the person running the test confirm the results by tapping or shaking the phone, so that it stays in the patient’s field of vision. The phone also checks the ambient light in case that’s affecting the test. Here again, it’s understanding that testing can happen anywhere – a school, outdoors, under a tree. It’s designing a product for where it will be used, not where the developer’s working.
So many mobile health apps try to turn patients into clinicians, giving them numbers and charts and expecting them to learn what they mean. As Andrew says, who other than an ophthalmologist really understands what it means when we’re told we have 60/60 vision? They’ve developed their Sightsim app which splits the screen in half. On one side it shows what the patient’s current vision is like; on the other what it would look like after treatment. It’s a dramatic indication of what the change could be, not just for the patient, but for parents, teachers, friends and spouses who will need to explain and support the patient through the process. It doesn’t display an abstract number, instead it demonstrates a very visible goal.
By using the eye tracking functionality of recent smartphones, the field worker can also perform field measurements to test peripheral vision, and even assess the sight of babies from a few days old.
The most impressive innovation is a low cost lens assembly which can be fitted to the phone to give a retinal scan. Using image stitching it provides a full picture of the retina, comparable to one taken using equipment costing tens of thousands of pounds. As Andrew says in his talks, seeing the optic nerve is the start of looking into the brain.
Data can be streamed to a specialist, or if there’s no connection, stored for future transmission. The system stores the patient’s details and location (many have no meaningful address), along with the details of a friend or community figure who can pass on messages, understanding that sending a text message to someone with impaired vision may not be particularly useful. In tests they’ve conducted, it appears that their £60 system produces comparable data to today’s clinical devices which cost tens of thousands of pounds and which are only practical in a hospital environment.
There’s a history of glaucoma in my family, so I have an annual check-up during which I feel rather like a laboratory rat, getting shuffled from room to room as my eyes are checked with a battery of devices that were designed for doctors and not patients. It’s not just in hospitals. The three opticians on my street are engaged in an ophthalmological arms race to have the latest and most patient unfriendly pieces of technology that goes ping, as if they believe that scaring the patient is the route to customer acquisition. I would love to see them moving to the Peekvision product. It’s far less intimidating than the equipment they currently use and would give me, the patient, far more useful information. With the money they would save, they might even consider donating to deserving causes like Peekvision.
I can’t really do justice to Andrew’s talk – this is just a short taster. It deserves to be seen by anyone designing a mobile health product, wherever it’s planned to be used. It’s all about being practical and putting the patient and their support community at the centre of the development process. It’s best summed up as the realisation that we don’t need health apps; we certainly don’t need medical apps; we need people apps.
You’ll get more than £60 of practical developer insight from listening to his talk. It’s a wonderful combination of pragmatism and vision. I had no hesitation in contributing to the IndieGoGo campaign and I’d urge you to join me. If your company donates to charity for Christmas, consider donating to Peekvision; if you’re in tech your customers will appreciate the gesture. It would be wonderful if this were to become one of the most oversubscribed projects ever on IndieGoGo. It certainly deserves to be.