Merlin Neurotech @ Queen’s University just submitted their final project for the 2018 student club competition hosted by NeuroTechX.
Our project is called WizardHat, a modular and flexible python API for the streaming and manipulating of EEG data for BCI development.
To learn more, check out our youtube video, and to test it out yourself, here is our project’s github.
Merlin will spend the next few months refining and improving our WizardHat platform for distributed use by our current and future members. We will be assembling a beginner’s guide for those interested in joining our club and developing brain-computer interfaces.
If you are interested in early-involvement and taking on an executive role in club leadership, send us an email through our contact us page. Other interested members will have the opportunity to apply in the Fall semester of 2018. Keep a lookout for details about the information session around that time.
Merlin Neurotech welcomed several new members this month. In order to teach everyone the basics of Brain-Computer Interfaces, we have split up into two teams where each team’s goal is to build their own binary classifier using the Muse EEG device. We hope to have a working demo by the end of the month. Stay tuned.
We ran another P300 experiment- this time using an auditory Oddball paradigm. After collecting the data, we ran it through some cookie-cutter classification pipelines to get a feel for the trial-by-trial SNR (signal-to-noise ratio) associated with the P300. All of the methods demonstrated a pretty low AUC (area under the curve), which means that on a trial-by-trial basis the classification isn’t very accurate, but we’re still above chance-level! We’re going to keep playing with some of our recorded data and try to implement some of our own algorithms.
An event-related potential (ERP) is a detectable brain signal related to the occurrence of a stimulus, cognitive, or motor event. P300 wave is an ERP related to decision making in the brain. It can be detected with an EEG using a simple Oddball paradigm where a target observes a screen which alternates between the target (stimulus) and non-target images.
For our first experiment, we replicated the Oddball paradigm using sample code from Alexandre Barachant’s muse-lsl workshop. Our results show the presence of the P300 wave in the TP9 and TP10 electrodes located behind the ears on the Muse brain sensing headband.
Here we can see the average signal from each of the four electrodes over all trials. For electrode TP9 and TP10 there is a clear detection of the P300 wave evident by the difference of target vs non-target potentials.
We are happy to announce that we had just received our first EEG headset through our partnership with Muse and NeuroTechX. Next, we will be carrying out some basic EEG experiments in order to learn how to analyze and classify EEG data for our future BCI projects. Stay tuned for our next blog post.
We are excited to announce the formation of a new student club at Queen’s University: the Merlin Neurotechnology Club. We’re a group of undergraduate and graduate students who are passionate about the development of a neurotechnology community at Queen’s.
What is neurotechnology?
Neurotechnology is a vast interdisciplinary field that encompasses any kind of technological development involving the brain. This includes imaging, stimulation, invasive devices, and even genetic and cellular techniques. Our club will focus on BCIs: brain-computer interfaces. We’re really excited about developing technology that allows direct communication between brains and computers. Yep. We want to control computers with the power of thought, and we’re not alone! In the past 6 months, both Elon Musk and Mark Zuckerberg have announced major BCI development projects, leading to an industry explosion and a race to design the most functional neurotechnology.
Why is it called “Merlin” Neurotech?
The founders of this club were heavily influenced by this awesome write-up by Wait But Why. The author jokes that the goal of Elon Musk’s Neuralink initiative is to build a “wizard hat for the brain,” and we’ve decided that our goals are no less lofty. As powerful as he is, “Gandalf Neurotech” just didn’t have the same ring to it.
How do you plan to develop this club?
Fortunately for us, a blueprint exists. NeuroTechX is a non-profit organization that aims to provide a framework for the development of an international neurotechnology community. They are working to provide neurotechnology enthusiasts from all disciplines and all corners of the globe with the knowledge, resources, and connections to dive into this exciting field. They have a very active online community, open-source coding and development platforms, and industry partnerships- there are currently NeuroTechX chapters in 18 cities and 11 universities worldwide! They will be providing us with commercial EEG (electroencephalography) devices to begin recording brain signals; we plan to start by reproducing some simple EEG experiments, then we’ll move on to more exciting projects as our skills develop.
How can I get involved?
We’re currently still in the process of officially ratifying our club with Queen’s, and while we get the ball rolling we’re hoping to keep things small. That said, if you’re keen to be a part of our club, or if you have any questions about what we do, shoot us an email at firstname.lastname@example.org. We’ll also be posting updates on our projects on this blog as we start wizard hat development, so check back here for more info!