The heart of this breakthrough lies in a sophisticated artificial intelligence system designed around a feature-augmented transformer model. Unlike traditional classification systems, this model leverages vast amounts of labeled data to detect higher-level, goal-directed behaviors such as cooking, working, socializing, or running errands by analyzing sensor signals during everyday life outside laboratory constraints. The algorithm achieves this through advanced pattern recognition that synthesizes diverse smartwatch sensor inputs over time, thereby accurately mapping complex behavioral patterns in real-world settings.

Over a span of eight years, the WSU research team gathered smartwatch data from more than 500 participants involved in various studies. Participants were prompted randomly throughout the day to self-report their ongoing activity from a predefined list comprising 12 categories, including sleeping, traveling, eating, and relaxing, among others. This unique approach yielded an unprecedented dataset exceeding 32 million labeled data points, each representing a one-minute segment of recorded activity paired with participant-reported labels. Such an expansive and richly annotated dataset provided a fertile ground for training the transformer model and testing its ability to generalize across diverse individuals.

What distinguishes this work is not only its scale but its capacity to recognize high-level functional activities essential for daily living, which are often reflective of a person’s independence and well-being. Clinical practitioners have long faced challenges in remotely assessing how individuals, especially the elderly or chronically ill, manage routine but critical activities such as handling finances, preparing meals, or performing self-care tasks. Traditional clinical visits offer limited snapshots, and prior use of wearable devices mostly focused on simple motion detection. The model developed by WSU researchers bridges this gap by enabling continuous, automated assessment of functional behaviors through passive monitoring.

The technical backbone of the system centers on cutting-edge transformer architectures, which have transformed natural language processing and signal analysis by capturing contextual relationships over time more effectively than prior recurrent or convolutional models. By augmenting features derived from multi-modal smartwatch signals, the transformer model aggregates temporal dependencies and subtle signal variations associated with different activities. This composite analysis enriches the recognizability of goal-directed behaviors despite the inherent noise and variability present in uncontrolled, “in-the-wild” data collection settings.

Achieving an overall activity recognition accuracy of nearly 78%, the system marks a notable advance in wearable sensing technology. This level of precision allows for actionable insights into patterns of behavior that correlate with cognitive and physical health metrics. For example, decreased frequency or altered sequences of specific activities can serve as early indicators of cognitive decline or diminished physical capacity. Consequently, the model offers potential pathways for proactive healthcare interventions that preserve independence and improve quality of life for at-risk populations.

Beyond clinical applications, the research lays foundational groundwork for the integration of human-centered artificial intelligence in digital health ecosystems. By capturing detailed behavioral signatures unobtrusively, such frameworks may support remote caregiving, personalized therapy regimens, and even automated clinical diagnostics once coupled with electronic health records and genetic information. Furthermore, making the anonymized dataset and modeling methods publicly available opens doors for the broader scientific community to innovate around activity recognition, behavior prediction, and human health analytics.

Lead researcher Diane Cook, a Regents Professor in WSU’s School of Electrical Engineering and Computer Science, emphasizes the societal importance of this work. She highlights how understanding an individual’s capability to perform critical activities—like bathing, feeding oneself, or managing finances—is fundamental to assessing independence. “If we can describe a person’s behavior in categories that are well recognized,” Cook notes, “we can start to talk about their behavior patterns and changes, which in turn relate to measures of cognitive health and functional independence.”

The project’s funding by the National Institutes of Health underscores the biomedical community’s recognition of wearable AI’s transformative potential. The interdisciplinary nature of the research spans computer science, behavioral health, and gerontology, reflecting a growing emphasis on data-driven precision medicine. By innovating at the intersection of robust AI modeling and real-world human activity monitoring, the WSU team exemplifies how wearable sensors can move beyond step counting to become critical tools in managing complex health trajectories.

Moving forward, researchers aim to refine the model’s capabilities, exploring automated clinical diagnosis and investigating associations between activity patterns, genetics, and environmental variables. Such studies could enrich personalized medicine by contextualizing behavioral data within an individual’s unique biological and socio-economic framework. Moreover, the deployment of this technology in everyday consumer smartwatches and healthcare settings promises to democratize access to health monitoring, empowering individuals and clinicians alike with continuous, actionable insights.

In summary, the groundbreaking work by Washington State University researchers represents a significant leap in translating raw wearable smartwatch data into meaningful, high-level assessments of human activity in naturalistic environments. Through an innovative transformer-based AI approach and an extensive, meticulously curated dataset, this system offers a promising new avenue for remote health monitoring, cognitive assessment, and functional independence evaluation. As wearable tech becomes ubiquitous, such advances pave the way for a future where continuous health insights are seamlessly integrated into daily life, potentially reducing healthcare costs and enhancing well-being worldwide.

Subject of Research: Functional activity recognition using AI from real-world smartwatch data to assess cognitive and physical health.

Article Title: Feature-Augmented Transformer Model to Recognize Functional Activities from in-the-wild Smartwatch Data

News Publication Date: 4-Jul-2025

Web References:
https://ieeexplore.ieee.org/document/11071689
http://dx.doi.org/10.1109/JBHI.2025.3586074

References: Washington State University research team; National Institutes of Health funding

Keywords: wearable technology, smartwatch data, activity recognition, transformer AI, cognitive health assessment, functional independence, digital health, behavioral monitoring, machine learning, real-world data, healthcare innovation

Between April 2020 and December 2022, over half a million people in Germany accessed the Corona Data Donation App, a pioneering initiative aimed at gathering real-time health data from users about their physical metrics, including resting heart rates and daily step counts. Of those, more than 120,000 poured in their daily health stats, revealing how their bodies responded to and recovered from COVID-19. Researchers, spearheaded by Katharina Ledebur from the Complexity Science Hub (CSH), were able to finely analyze the metrics in 15-minute increments, creating an unprecedented comprehensive data set for examination.

Data analysis presented significant differences in activity and cardiovascular metrics between groups of patients categorized by symptom persistence. Specifically, individuals reporting enduring symptoms took an average of only 5,075 steps daily in the three weeks leading to their COVID-19 diagnosis. This figure starkly contrasts the 8,105 daily steps observed in patients recovering without persistent symptoms, illustrating an alarming decrease in physical activity correlated with the severity of long-term symptoms such as exhaustion and breathlessness.

The resting heart rate presented equally compelling findings. Prior to infection, patients who later developed long-lasting symptoms exhibited an average resting heart rate that was 2.37 beats per minute higher than individuals who recovered promptly. Additionally, the persistent symptoms group experienced a significant and prolonged incidence of bradycardia—a slowing of the heart rate—lasting up to 18 days post-infection. The ramifications of these findings underscore how the body may endure physiological changes long after the virus has cleared.

Despite these concerning variations, researchers noted a surprising trend: both groups ultimately saw their resting heart rates and daily step counts return to pre-infection averages. This finding suggests an intriguing resilience in cardiovascular and physical activity metrics, hinting that individuals, even those plagued by prolonged COVID-19 symptoms, could regain certain facets of their health over time.

When categorizing the observations chronologically, researchers segmented the data into four distinct phases: the pre-infection period, the acute phase (the first four weeks following a positive COVID-19 test), the subacute phase (spanning weeks five to twelve post-infection), and the post-acute phase (which commences beyond twelve weeks). By doing so, the study effectively delineated the trajectory of symptoms and physiological responses over time, providing a clearer picture of the lingering implications of the COVID-19 virus on health.

Further analysis of the participants’ monthly and weekly questionnaires highlighted ongoing symptoms, bringing attention to the fact that only two out of eleven reported symptoms—shortness of breath and fatigue—turned out to be persistent. This finding underlines the necessity for ongoing research to better understand the range of symptoms tied to long COVID, as these two particular symptoms were notably prevalent among 2.6% and 10.4% of SARS-CoV-2-positive individuals, respectively.

The complicated interplay between health metrics and lingering symptoms raises vital questions about underlying health factors and lifestyle dynamics prior to infection. Certain individuals appeared to be more susceptible to enduring symptoms, potentially due to lower fitness levels or preexisting health conditions. It is crucial to recognize, as emphasized by Ledebur, that elevated resting heart rates and reduced physical activity do not exist in isolation as sole causes of long-lasting symptoms. Rather, these findings warrant a comprehensive understanding of individual health profiles and tailored protective strategies for those at heightened risk of developing long COVID-related consequences.

The availability of high-resolution smartwatch data empowers researchers to conduct individual-level analyses that were previously hindered by a lack of baseline metrics prior to infection. This comparative strength allows for a more nuanced understanding of how significant health changes manifest post-infection, accentuating the transformative potential of wearable technology in unveiling hidden trends in public health.

Nevertheless, as striking as these findings may appear, it’s critical to acknowledge their boundaries. The demographic representation of participants skews towards males while notably underrepresenting adolescents and individuals aged 65 and older, populations generally at heightened risk during viral outbreaks. Additionally, it should be considered that those who own and regularly utilize fitness tracking devices may inherently possess a more health-conscious mindset, potentially skewing the data toward more favorable health outcomes.

Ultimately, this research serves as a potent reminder of the latent complexities intertwined with the COVID-19 pandemic. It reflects an urgent need for continued investigation into how we can leverage technology and real-world data to not only grasp the full spectrum of long COVID symptoms and their effects on health but also inform more effective interventions and health strategies for affected populations.

As the public continues to grapple with the implications of the COVID-19 pandemic, the insights drawn from these comprehensive analyses provide a foundation for future research, Texas findings forward, and a call to action for researchers and healthcare professionals to pay increased attention to the follow-up care of COVID-19 patients, particularly considering the possibility of enduring health consequences that may emerge from this multi-faceted illness.

The robust data gathered through initiatives like the Corona Data Donation App affirm the power of digital health surveillance in understanding large-scale health phenomena. This technology’s potential could reshape how we approach not only infectious diseases but public health initiatives overall, paving the way for a future that can use data-driven insights to improve individual and collective health outcomes.

In conclusion, the study encapsulates both a new understanding of the ramifications of COVID-19 on long-term health and an optimistic lens of resilience reflecting the body’s capacity to recover. As scientists continue to investigate the lingering effects of this virus, these findings serve as a crucial stepping stone to understanding the importance of ongoing support and monitoring for those experiencing long COVID symptoms.

—

Subject of Research: People
Article Title: Wearable data reveals distinct characteristics of individuals with persistent symptoms after a SARS-CoV-2 infection
News Publication Date: 19-Mar-2025
Web References: N/A
References: N/A
Image Credits: Complexity Science Hub
Keywords: Long Covid, Digital data, Acute infections, SARS CoV 2, Respiration, Heart rate, Risk factors