A science-backed environmental migraine risk monitoring card for Home Assistant. Tracks 9 weather and air quality factors known to trigger migraines and displays a composite risk score with a visual gauge.
Gamified migraine tracker PWA — Track sleep, water, attacks & triggers with XP, levels, quizzes, and rule-based insights. Built with React + Express + SQLite.
MigraineMamba predicts migraine attacks 24 hours in advance with 78% accuracy using Mamba (Selective State Space Model) architecture with self-supervised learning (SSL). The system personalizes over time through a 3-phase journey: foundation model (days 1-14), generic temporal patterns (days 15-30), and personalized fine-tuning (days 31+).
Portfolio of clinical data science projects focused on reproducible analytics, risk prediction, and clinically meaningful modeling across structured healthcare datasets.
A companion app for a speculative wearable biosensor designed to detect the migraine prodrome window before an attack begins.
Learns your personal pre-migraine pattern from your own Apple Health data (heart rate, HRV, respiratory rate, sleep, activity) tracked around the clock, sharpening as you log more headaches. Daily 3-stage status: all clear, early warning building, or strong match — attack likely within ~48h. Self-tracking tool, not a medical device.
Track migraine risk in Home Assistant with a weather and air quality card that scores 9 trigger factors and shows a clear gauge
To associate your repository with the migraine-prediction topic, visit your repo's landing page and select "manage topics."