The is a global curriculum and competition that challenges students to apply their skills and creativity to tackle today’s most pressing social and environmental issues. The AI 100 brief, sponsored by Philips asked "How might we use AI to support people to reach a happy, meaningful and productive one hundred year life?"
Product Design and Technology student, Lalith Sanathi, responded with his proposal for 'Insight' which is a simple, easy to use retinal camera. The eyepiece is integrated with the body, and has a subtle curve to aid in comfort. The user simply looks in to the camera. follows the on screen instructions, aligns their eye with the dot, and presses the shutter button.
Lalith explains that:
"There were a wide variety of briefs set but I was immediately drawn to the AI 100 one, which asked for designers to conceptualise a way to use AI to progress towards a more productive and high quality 100 year lifespan. I was already looking at a possible Alzheimer’s focus for my final year project, so alongside my proclivity for the AI aspect, this made the choice of brief a no brainer.
The core problem that I wanted to tackle was the general lack of understanding we have of Alzheimer’s Disease, both as individuals and as a society. One way to remedy this is through further development and research into early detection. However, current methods are inaccurate, not very early, and uncommon.
My response to the brief was a service and product proposal that aimed to make early detection of Alzheimer’s much more likely, allowing both a better quality of life for sufferers, as well as a wider pool of people and data for use in research. The concept (Insight) consists of a smart camera that allows users to scan their eye for possible signs of Alzheimer’s which can cause macular degeneration years before any noticeable symptoms. The camera pairs with the user’s phone and uses computer vision to analyse the photo. It then flags up any signs of the disease and prompts the user to book an appointment with their physician for further testing. The app also provides support and guidance throughout the user’s journey to help them understand and cope with their disease.
One of the challenges I felt was important to tackle was the routine building aspect. The camera is only as good as the data it’s given, but regular monthly scans are hard to remember to do, and reminders and other methods of trying to force a scan are inelegant and can be annoying. Insight solves this by detecting when the user is due for a scan, and simply tilting itself to stand off kilter by raising the camera shutter button. This is just enough of a nudge to prompt the user to notice and “fix” the issue by taking a photo.
Working on this project allowed me to experiment with new concepts, ideas, skills, and frameworks. For example, to build a proof of concept, I trained a neural network on open source retinal scans to accurately assess whether an eye is healthy or not, as well as differentiate between diseases."
