MigraineMinder is an intelligent migraine management assistant that bridges neurotechnology and environmental control. Using real-time EEG signals from the Muse 2 headband, the system detects early neurological signatures of migraine onset and automatically adjusts ambient lighting or screen brightness via an Arduino-controlled light module.

When a potential migraine episode is detected, MigraineMinder gently dims the environment, logs the event, and prompts the user to record contextual causes — stress, noise, caffeine, or the inevitable group project panic.

As neurologist Oliver Sacks wrote in Migraine,

“The migraine is not a disease of the brain, but a disorder of energy, of control, of balance.”
MigraineMinder aims to restore that balance — one photon at a time.

• Flask – for the dashboard and migraine event logging
• Python (Muse SDK) – for real-time EEG signal acquisition and anomaly detection
• Arduino (C / C++) – to control LED brightness and communicate with sensors
• Machine Learning Models – Deep learning for migraine prediction
• CSS / JS – for the interactive migraine diary interface

• Muse 2 Headband – EEG, heart rate, motion tracking
• Arduino Board – bridges software with the environment
• LDR Sensor – monitors ambient light
• OLED Display – shows live brain and light status
• Vibrating Motor – gives subtle tactile feedback during pre-migraine warnings
• DHT11 Sensor – records temperature and humidity (potential triggers)

The spark came from one teammate’s grandmother, who endures frequent migraines that make everyday light feel like a flashbang. We wanted to make something that reacts faster than a person can — a little guardian that listens to your brain before the pain hits.

It also doubles as a savior for university students who can finally say,

“Sorry, Professor — my brainwaves literally refused to let me finish the assignment.”

• 🧠 Detects migraine patterns from EEG and physiological data (stress, HRV, focus drop).
• 💡 Adapts the environment: dims lights automatically or prompts users to reduce screen brightness.
• 🗒️ Logs triggers: prompts users to tag the cause (noise, fatigue, weather, coursework).
• 📊 Visualizes data: shows EEG trends, trigger frequency, and environment correlation.
• 🤖 Personalizes recommendations: over time, MigraineMinder learns which conditions precede migraines.

• Distinguishing genuine migraine signatures from everyday stress signals (turns out, debugging at 3 a.m. looks a lot like a migraine).
• Synchronizing Muse 2’s Bluetooth data stream with Arduino’s serial communication.
• Managing false positives — we accidentally dimmed the lab lights every time someone yawned.
• Balancing data privacy with usability for medical-adjacent data.

• Successfully built an EEG-to-light feedback loop using Muse 2 + Arduino.
• Developed a clean Flask-based dashboard to record and visualize migraine episodes.
• Created a system that detects early migraine trends with promising accuracy in pilot testing.
• Built a user-friendly front-end interface that looks cool even at 20 % brightness.

• The human brain is the ultimate noisy dataset — and patience is a debugging skill.
• Integrating biosensors with real-world IoT control is both thrilling and humbling.
• Empathy can inspire powerful design — when you build for one person’s pain, you often help many.

• Expanding detection accuracy with machine learning trained on diverse EEG data.
• Integrating with smart-home ecosystems (Alexa, Philips Hue) for seamless light control.
• Building a mobile companion app for quick trigger logging and real-time alerts.
• Collaborating with neurologists to test MigraineMinder as a clinical support tool.

In the words of Oliver Sacks,

“Migraine is a kind of electrical storm of the nervous system.”

MigraineMinder doesn't try to stop the storm — it just knows when to turn down the lights.