So, this week started out with a surprise – I got a chance to test my system with a super cool 12 year old girl – without ADHD 🙂
I was super exited and motivated to take my system out of the lab and test it in a real world setting! My ADHD Monitoring System works in the lab – but that doesn’t always mean that it works in a real context. Luckily, my system actually did work as expected – almost 🙂 Of course it was important that my system worked on a practical level, but the most interesting part was whether the system could detect a difference in “activity behaviour” or nor. As seen on the output image above, it seems like the system can detect a difference between a non-ADHD child (red graph) and a child with ADHD (green graph). Please note that both the red and green graphs are from my study with the same girl – she were just being instructed / encouraged to move more in the later experiment.
To sum up, I was very glad that my system worked and that it seems that the (almost raw) data has potential to detect the difference between ADHD children and non-ADHD children. And if that is possible, the next step is to teach the system to detect when the child is beginning to lose their concentration based on their activity level / patterns.
I also wrote a little more on my thesis – I had hoped that my advisor would have had the time to give some feedback on my writing so far, but he canceled our meeting TWICE ! Unfortunately, I don’t feel I have a clear direction for my thesis right now, so I really hope that I can get a meeting with him soon..
I will end this status update with a few lessons learned from my experiment:
- Deactivate the power-saving function before an experiment! (argh!)
- Most hardware / sensors such as EEG headsets and Heart Rate Belts aren’t made to fit perfectly to a child – hacks are needed!
- And Yes, a Samsung Galaxy S3 is a very big smartphone to put in a child’s pocket.
- Don’t spend a lot of money buying a wristband, if the one you made yourself actually is the one the child prefers #MoneySavingTip