Quadrapelgic cerebral palsy, abbreviated as CP, is a movement and cognitive disorder that results in the disruption of brain connectivity in up to three out of every 1000 live births [1]. In part due to their inability to move their limbs, those impacted by CP tend to suffer from a lack of social autonomy. This may contribute to the higher rates of anxiety and depression seen in this population [2]. In this way, developing interactive experiences for those with CP to practice relaxation may alleviate the mental health burden on these individuals.

[1] Vitrikas K., Dalton H., Breish D. (2020). Cerebral Palsy: An Overview. American Family Physician. 101(4):213-220

[2] Smith KJ, Peterson MD, O’Connell NE, Victor C., Liverani S., Anokye N., Ryan JM (2019). Risk of Depression and Anxiety in Adults With Cerebral Palsy. JAMA Neurology. 76(3): 294-300

What it does

BrainRunner is a 3D endless running game that implements generic Brain-to-Browser software. Using a commercial brain-computer interface device, BrainRunner allows for players to control on-screen movements with live-streamed brain activity, receive feedback on their relaxation level, and compete against friends, family, and anyone else in the world. During gameplay, the goal is to collect as many coins as possible before the end of each round. As you become more relaxed, it becomes easier to collect coins.

How we built it

Developed in consultation with the BCI4Kids program, BrainRunner is one of many games that could support improved quality of life for those impacted by CP. Enabled by the Brains@Play software library that our team helped to expand during HackSC, we hope that this interactive game promotes other developers to work towards a brain-accessible web for individuals with severe disabilities.

Challenges we ran into

BCI is a relatively new field and hence lacks appropriate tools for developers. As a result, we had to make use of new libraries like BCI.js and also our very own library BrainsAtPlay.js to create a web portal for developers. We faced numerous challenges concerning visualizations, like integrating P5.js and Three.js which have different workflows and not-so-good documentations. Plus, since this was a one-of-its-kind application, there was no reference available and we had to use our own ideas and intuitions to see what worked best.

Accomplishments that we're proud of

Finally, since there were no data or pre-trained classifiers available, we had to create our own data acquisition and classifier training pipeline. Although the task was exhausting, we believe we have laid a great foundation for creators who in the near future can create their own apps, obtain their own data and train their own classifiers. This will in turn provide accessibility to the disabled to carry out a wide range of tasks using the power of the web.

What we learned

Also, we decided to provide the ability of multiplayer interactions in the apps which led to the issue of creating a robust, near-real-time websocket to transmit data. Finally, since there were no data or pre-trained classifiers available, we had to create our own data acquisition and classifier training pipeline.

What's next for BrainRunner

We are planning to make the Brains@Play portal and library publicly available so developers can create and post their own apps for children suffering from cerebral palsy.

Built With

Share this project: