Wearable Sensors and Machine Learning: Insights into Depression, Anxiety, and Emotional States and Changes

Mental health disorders, such as depression and anxiety, significantly impact individuals worldwide, yet timely diagnosis and intervention remain challenging due to the subjective nature of traditional methods. Wearable technology, including devices such as smartwatches and smartphones, combined with machine learning, offers new opportunities to monitor and predict mental health conditions. This paper provides a comprehensive review of studies published between 2013 and 2023, focusing on the application of wearable devices and machine learning to predict emotional states and changes. We summarize commonly used machine learning methods, such as support vector machines and random forests, alongside key features extracted from wearable data, including physiological indicators and behavioral patterns. Unlike previous reviews, this work emphasizes the underexplored area of emotional changes, identifying gaps in current research and proposing future directions. The findings aim to inspire innovative approaches to the monitoring and intervention of mental health.