Technology with attitude

Brainihack 2015 hackathon roundup

2

Last week I was in Tel Aviv, Israel, for the BrainTech Conference and Brainihack, a BCI focused hackathon. A weekend long event where about a hundred people from different backgrounds come to work in team on a fun project related to brain-computer interfaces.These hackathons are getting more and more popular because  (1) they are fun, (2) people get to work very intensely on a cool project of their choosing, (3) very short, (4) you meet a bunch of people with crazy ideas. As the time passes by, the stress kicks in and people produce their best work sometime leading to something amazing, sometime leading nowhere. Hackathons are not about the product itself, but about the journey, the people and the ideas. No one expect a perfect product, ready to be shipped, in a weekend.

Hackathons are a very good tool to compare yourself with other peers, skill-wise. Self-confidence arise from knowing what you can & can’t do and the younger you realize your potential and limitation, the better.

In the book Talent is Overrated, Goeff Colvin talks about the multiplier effect and how realizing that you are slightly better than others in one field, can lead to tremendous results down the road. Young talents often think that everyone else is better and that the world is filled with “super-talented-can-do-everything-superheroes”, because these are the success stories we hear and see, but the reality of course, is the opposite. The world is filled with “super-average-cant-do-much-ordinary-people”. Don’t get me wrong here, I don’t want to be pessimistic so let me give you some examples.

IMG_0752

When you are young you think that adults and parents know everything. As you grow old, you quickly learn that they don’t. Lifelong trials and errors, called experience, nothing more. As you get to university you think: wow this is major league, everything must be perfect. The professors, the classes, the exams, the labs, etc. Then you become a grad student and realize it’s not. When you’re studying engineering and looking for a job, you feel like you are under-qualified for every interesting offer, that the industry must be perfect. You look at products like your laptop and tell yourself, these guys are geniuses and made a perfect product. Then you get the job and realize that products get shipped on a Friday night with minor defects to meet the deadlines. That the design is far from perfect, but good enough to ship and get paid. You realize that there are very talented people in that company, but they are a minority (unless you work for companies like Google and Apple, but statistically speaking, these are not the majority either.) Realizing that the world is not perfect and that for the first time in human history, we all have access to enough information to master any field in a relatively short period of time is priceless.

Chicken or the egg  (talent or practice ?)

So what is that multiplier effect? It’s the chicken and egg problem, but for talent. Do you do something because you are good at it or you are good at something because you do it? It has to start somewhere right? Doing something and realizing that you can do it slightly better than your peers will bring attention (from parents, peers, teachers, recruiters, etc.) and praise that will fuel your desire to practice more in this field in order to get better and bring more attention and praise. The loop goes on, creating a self-feeding ‘passion’. Combine that with the dopamine being released when overcoming challenges and you’ve just hacked your brain to becoming really good at “that something”, by using its own powerful chemicals: dopamine and serotonin.

Wow, I got carried away! Probably the excess of coffee.

Back to the Hackathon…

I was amazed by the quality of people. Undergrad & grad students, professionals,  professors from different universities to help teams and even Ariel Garten, CEO of InteraXon, Conor Russomanno, CEO of OpenBCI and Stephen Dunne from NeuroElectrics.

Hamutal Meridor is the one behind Brainihack and she did an amazing job, this year again, giving the chance to a hundred of people to play with BCI devices and either learn the hard lesson of their limitation for the first time or use their knowledge in the field to work on more advanced projects. “It’s a little messy, but it always is and always work out great.” said Hamutal, during the intro. Hackathons are organic and people participating to these events are awesome and passionate. Not sure you can really screw up a hackathon.

Hamutal Meridor, the brain behind Brainihack
Hamutal Meridor, the brain behind Brainihack

We had some cool projects ranging from research grade to cool hacks. The favorite devices were OpenBCI and Muse, but some teams also used the Enobio, the g.Tec, the Neurosky, Emotiv EPOC, Oculus Rift and even a 3D printer! Nathan Intrator, professor at Tel Aviv University and Brown University also brought “Neurosteer” a new EEG headband he’s working on (exciting!).

For the full list of teams and projects, see Hamutal’s recap available here.

Emochat, the winning project, used the Neurosteer device to make a BCI-chat where you can see the other person’s emotions as you exchange, based on their brain activity. No more misunderstanding because you don’t see facial expression or don’t hear voice tone, now you have a direct access to the other person’s brain! (almost :p). We also saw a “Date my Brain” project, finding the perfect match for you based on your brainwaves. A multiplayer game supporting all the BCI devices where each player connects is own device on his computer and join the online party to fly a bee. A project trying to detect a bike rider’s intention to turn left or right and automatically switch on the appropriate flasher without having to lift a finger! Another one indicating if you recognize the person on a picture to create a ‘lie detector / tool for conviction’.

My 2 favorite ones were the maze project with SSVEP using the OpenBCI and the Lego Mindstorm EV3 being controlled with Motor Imagery using the g.Tec headset.

SSVEP & OpenBCI (from Blue GSD)

The team Blue GSD 3D printed a maze with two micro servo motors controlled by an Arduino. With these 2 degrees of freedom you can incline the maze in order to move the ball across. They mapped these 2 degrees of freedom with the Alpha peak and SSVEP triggered by a cellphone screen flashing at a specific frequency (they had good result with 16Hz). Check out their project & code here. This is what I call “hacking”! The day after the hackathon they were featured on Hackaday.com

SSVEP
Team “Blue GSD” created a 3D printed maze with two micro servo motors controlled by an Arduino

Motor Imagery & g.Tec (from Mind over mindstorms)

Two undergrad students working in a research lab came with their professor and worked on their motor imagery BCI project to control a Lego Mindstorm robot. Sitting still and thinking about moving her left hand for the robot to turn left and thinking about moving the right hand for the robot to turn right. Simple? The idea certainly is, but the underlying neuroscience and technology isn’t. Of course this research level project wasn’t done during the weekend, but it was nice to see that kind of project at the hackathon. It definitely inspired more than one hacker to believe that the future of neuro-technology is incredible.

Mindstorm
Lego Mindstorm EV3 being controlled with Motor Imagery using the g.Tec headset

Conor from OpenBCI and Ariel from InteraXon were incredibly generous by helping teams. Conor stayed there and worked hard to help every OpenBCI team to achieve their goals the whole Hackathon. Luckily I could grab him for a quick video interview that will be available soon. (need to be edited, because we got interrupted by hackers requesting his help).

Ariel had Tom Hobson from Toronto to wake up at 5 in the morning the give a remote workshops and Q&A on the Muse through hangouts. InteraXon and OpenBCI are genuine BCI community oriented companies. Stephen Dunne from Neuroelectrics was a little less solicited as one could expect in a hackathon, but very generous with those requiring his help.

Ariel
Ariel Garten with Tom Hobson from Toronto, giving remote workshop and Q&A on the Muse headband
Conor
Conor Russomanno worked hard to help every OpenBCI team to achieve their goals

BCIs being BCIs, the live demos were not successful for most. BCIs are unlikely to work with a very high accuracy for live demos simply because you are stressed out and your calibration/machine learning algorithm were not trained with that amount of stress. Also, we’re talking about controlling our brain in front of 100 people in real time.

I couldn’t end this article without mentioning the view… Autodesk offices in Tel Aviv offers an incredible view on the city and the see. Insane panoramic view, check out the photos!

As I write those lines, there is a brain/bci hackathon happening in UK called Hack the Brain UK. Neurogadget wishes them a weekend full of good hacks.

Brainihack already has another awesome BCI hackathon planned for May 2nd and 3rd in San-Francisco, prior to the Neurogaming conference which I can’t wait to cover with Melanie!

BCIMontréal (Montréal, Canada) is also organizing a BCI-related hackathon called “Your Body as an Input” in June (soon to be announced!)

The future of BCI is exciting ! If you are working on cool BCI projects, please share them with us on Facebook and Twitter !