Wearables could play an important role in monitoring brain health and detecting early signs of neurological or mental health issues. That is the conclusion of a study by researchers at the University of Geneva (UNIGE), who used artificial intelligence to analyze data collected from wearable devices. The results suggest that connected devices can help predict fluctuations in emotional and cognitive functioning.
Brain health, encompassing both cognitive abilities and emotional well-being, is increasingly recognized as a major public health challenge. According to the World Health Organization, more than one in three people worldwide lives with a neurological disorder, while more than half of the population will experience a mental disorder at some point in their lives.
Continuous monitoring of brain health
Even among healthy adults, brain health varies over time due to factors such as lifestyle, environment and daily routines. Being able to track these fluctuations could support earlier detection of problems and enable more personalized prevention strategies. Researchers at UNIGE therefore explored whether wearable technology could provide continuous, noninvasive monitoring of these changes. The research was recently published in npj Digital Medicine.
The study involved 88 participants aged between 45 and 77. Each volunteer used a dedicated smartphone app and wore a smartwatch over a period of ten months. The devices collected passive data without requiring participants to change their daily routines. These data included heart rate, physical activity levels, sleep patterns, weather conditions and air pollution exposure. In total, researchers analyzed 21 indicators.
Every three months, participants also contributed active data by completing questionnaires about their emotional state and performing cognitive tests. These results served as a benchmark for evaluating the predictive power of the AI models.
AI predicts emotional and cognitive fluctuations
After the data collection phase, researchers applied artificial intelligence to analyze the passive data and determine whether it could predict changes in participants’ cognitive and emotional health. “The goal was to see whether AI could forecast fluctuations in brain health based on data from wearable devices,” explains Igor Matias, doctoral researcher at the Research Institute for Statistics and Information Science at the Geneva School of Economics and Management and lead author of the study.
When compared with the results of the questionnaires and cognitive tests, the AI predictions showed an average error rate of 12.5 percent. According to the researchers, this level of accuracy indicates significant potential for wearables in the early detection of changes in brain health.
The analysis revealed that emotional states were the easiest to predict. AI models achieved error rates of around 5–10 percent when forecasting emotional questionnaire results. Predictions related to cognitive performance were less precise, with error rates ranging between 10 and 20 percent.
The researchers also identified which passive indicators were most relevant. For cognitive functioning, key factors included air pollution, weather conditions, daily heart rate and sleep variability. Emotional states were most strongly associated with weather conditions, sleep variability and heart rate during sleep.
Follow-up
The study was conducted as part of the Providemus alz research project and was supervised by Professor Katarzyna Wac and Professor Matthias Kliegel of the University of Geneva. A follow-up phase is already underway. Over the next two years, researchers will continue collecting wearable and smartphone data while examining which individual characteristics influence the performance of the AI models.
The aim is to better understand how such technologies could be applied in real-world healthcare settings. In the future, continuous monitoring via wearable devices could help clinicians identify early changes in brain health and support more personalized prevention and care strategies.
Wearables and health monitoring
Last year, researchers at the University of Missouri developed a starfish-inspired wearable device designed to improve heart monitoring during movement. Unlike traditional wearables that rely on a single contact point on the skin, the new design features five arms with sensors that measure both electrical and mechanical heart activity simultaneously.
This multi-point approach provides more stable and accurate data, even when the wearer is moving. Artificial intelligence analyzes the collected data, filtering motion-related noise and detecting cardiac irregularities with more than 90% accuracy. The device connects to a smartphone app, allowing users and healthcare professionals to monitor heart health remotely.
