Wearable data reveals brain health signals
Everyday wearable sensor data, combined with AI, can detect subtle changes in cognitive and emotional well-being [https://news-medical.net/news/20260312/Everyday-wearable-data-could-reveal-early-brain-health-signals.aspx]. This could identify early signs of stress or mental fatigue, paving the way for proactive wellness.
The recent study in *npj Digital Medicine* highlights the potential for everyday tech to track brain health at a population level. Researchers used consumer-grade wearables to develop models that quantify variations in daily cognition and affect. This offers a scalable way to detect subtle shifts in brain function long before clinical symptoms appear. The study, part of the Providemus alz project, followed 82 healthy adults for 10 months, gathering data from smartwatches and phones. Passive data included heart rate, sleep, activity, air quality, and weather. AI analysis of this data predicted cognitive and emotional states with an average error rate of just 12.5%. Emotional states proved easier to predict (5-10% error) than cognitive states (10-20%). Key predictors of cognition included air pollution, weather, heart rate, and sleep. For emotional states, weather, sleep, and heart rate during sleep were most influential. A follow-up study is underway to see how individual characteristics affect the AI models.
