Abstract
Introduction The landscape of personal health monitoring is rapidly transforming, driven by advances in wearable technology. These devices, designed to be worn comfortably on the body, can capture real-time physiological data often difficult to obtain in traditional clinical settings. What began as a simple device for activity trackers for walking steps, heart rate, and blood oxygen saturation has evolved to play a crucial role in monitoring more complex physiological parameters critical for chronic disease management and preventative health. The integration of artificial intelligence (AI) is further expanding the potential of wearable devices, enabling deeper analysis and predictive modeling of the recorded data. As AI-driven models become more refined, wearable technologies are poised to transform health monitoring, enabling early disease detection and personalized interventions. In many ways, we are at a pivotal moment in digital health, well-positioned to reimagine how healthcare is delivered, monitored, and personalized. This editorial discusses emerging clinical applications of wearable technologies across different life stages in patient care, the key limitations, and prospects in the rapidly advancing field. These include sleep monitoring in children, menstrual cycle tracking in reproductive-aged females, and cardiovascular monitoring in Parkinson's disease. Collectively, these studies highlight how continuous, real-time physiological data can uncover underlying subtle biological patterns and inform tailored interventions. Wearables in Pediatric Sleep Monitoring In children, particularly those with sensory sensitivities, the settling-down period (time between the end of bedtime routine and sleep onset) can be challenging. In a recent study, Hartman and colleagues investigated how wearable devices recorded data combined with machine learning models could help characterize the differences in the settling-down period in children with and without sensory sensitivities (SS/NSS) (Kocanaogullari, Akcakaya, Bendixen, Soehner, & Hartman, 2025). The study involved continuous two-week actigraphy data (measurement of rest and activity cycles) collection using the ActiGraph GT9X device worn on children with (n=17) and without (n=18) sensory sensitivities. Combining caregiver reports and the motor activity actigraphy data, the authors selected seven features and trained a machine learning model to distinguish between the SS and NSS groups with a high accuracy (84.1%), specificity (82.7%), and sensitivity (84.4%) (Figure 1A). This study highlights the potential of wearables in pediatric behavioral health and opens new avenues for personalized pediatric behavior interventions. However, limitations such as small cohort size, lack of diversity in the study groups, variability in wearable device placement, and reliance on subjective caregiver reports highlight the need to use larger and diverse sample cohorts, confirm observed activity differences, and apply them in intervention studies for children with sleep difficulties. Future studies should involve a larger and more diverse sample cohort and incorporate additional physiological parameters, e.g., heart rate variability and skin temperature, to provide a more comprehensive understanding of the physiological state during the settling-down period. Wearables in Advancing Reproductive Health through Cycle Tracking In reproductive-aged females, wearable technologies offer novel opportunities to track the menstrual cycle and optimize menstrual health management. Nasseri's group used a wearable device to understand female reproductive physiology and investigated menstrual cycle dynamics (Sides et al., 2023). In this study, physiological data were recorded using wearable sensors (Empatica E4 wristband) from healthy female individuals (n= 15; 18–40 years old) across 39 cycles (31 ovulating, 8 non-ovulating). The collected data included mean temperature, heart rate, inter-beat interval, and the mean tonic component of Electrodermal Activity (EDA). The authors used circular statistics to select key features for time series forecasting using the Autoregressive Integrated Moving Average (ARIMA) algorithm. This approach accurately identified and predicted menstrual cycle phases, distinguishing between ovulating and non-ovulating cycles (Figure 1B). The study contributes significantly to precise menstrual health management and personalised medical interventions based on distinct physiological patterns recorded by a wearable device. For instance, a reproductive-aged female undergoing medication while trying to conceive could monitor and adjust dosages in consultation with a physician according to their menstrual phase. Similarly, females with health issues linked to a specific menstrual phase, such as catamenial epilepsy, may benefit by adjusting the anti-seizure medication dosage based on their ovulatory status. The study has limitations, including a small sample size, data variability based on the wristband positioning, and the absence of blood hormonal levels to detect ovulation. Nonetheless, accurately identifying and predicting menstrual phases and distinguishing ovulating from non-ovulating cycles holds immense promise for precise menstrual health management and individualized medical interventions. Wearables in Cardiovascular Autonomic Monitoring in Neurodegenerative Disease In older adults with neurodegenerative conditions, managing autonomic dysfunction is particularly challenging. Berkebile`s group evaluated the feasibility of a wearable chest patch to monitor cardiovascular autonomic (CVA) function in Parkinson's disease (PD) patients undergoing levodopa treatment (Kehnemouyi, Coleman, & Tass, 2024). Levodopa, the primary medication for motor symptoms in Parkinson's disease and multiple system atrophy (MSA), can exacerbate cardiovascular autonomic (CVA) dysfunction due to its potential hypotensive effect. Monitoring CVA function is...