Inspiration
I remember visiting my uncle in the hospital after his stroke. Machines kept going ding, ding, and I wondered why his brain failed so quickly. That moment inspired me to explore brain resilience with math and CS.
What it does
This project models the brain as a network of regions connected by pathways and measures how efficiently information flows using the metric of Global Efficiency. By simulating a stroke removing the most connected hub region I show how overall communication in the brain drops, mirroring why hub damage often leads to severe outcomes. The project is designed as an educational tool to make brain resilience visible, but it also points toward future applications in stroke risk detection, treatment planning, and rehabilitation strategies.
How we built it
I built the project in Python on Google Colab, since my laptop crashed and Colab let me code fully in the cloud from my phone. Using NetworkX, I modeled the brain as a network of 20 regions connected by pathways. With NumPy, I ran calculations, and Pandas organized the results into clean tables. The core was measuring Global Efficiency, which reflects how well information flows across the brain. I first measured a healthy baseline, then simulated a stroke by removing the most connected hub node and recalculated efficiency. The difference revealed how losing a hub makes the brain less resilient.
Challenges we ran into
A key technical challenge was computing efficiency reliably on a phone within one day, after my laptop unexpectedly crashed. I had to figure out how to simulate resilience using only one core metrics,
Accomplishments that we're proud of
I am proud that I created a working simulation under my extreme constraints on time, no laptop, and coding entirely on my phone. I am also proud that I made the project simple and easily understandable.
What we learned
This project models the brain as a network and measures its resilience using Global Efficiency. By simulating random failures versus targeted hub lesions, it shows how damage in specific regions can collapse communication much faster explaining why some strokes are more devastating than others. The tool highlights critical nodes in the brain, making resilience measurable
What's next for NeuroNet-R
The next step is to expand the model with real brain atlas data, making the results more biologically accurate. I also plan to add additional metrics like clustering coefficient.
Log in or sign up for Devpost to join the conversation.