AdaptStress: Online Adaptive Learning for Interpretable and Personalized Stress Prediction Using Multivariate and Sparse Physiological Signals
arXiv:2602.18521v1 Announce Type: new Abstract: Continuous stress forecasting could potentially contribute to lifestyle interventions. This paper presents a novel, explainable, and individualized approach for stress prediction using physiological data from consumer-grade smartwatches. We develop a time series forecasting model that leverages multivariate features, including heart rate variability, activity patterns, and sleep metrics, to predict stress levels across 16 temporal horizons (History window: 3, 5, 7, 9 days; forecasting window: 1, 3, 5, 7 days). Our evaluation involves 16 participants monitored for 10-15 weeks. We evaluate our approach across 16 participants, comparing against state-of-the-art time series models (Informer, TimesNet, PatchTST) and traditional baselines (CNN, LSTM, CNN-LSTM) across multiple temporal horizons. Our model achieved performance with an MSE of 0.053, MAE of 0.190, and RMSE of 0.226 in optimal settings (5-day input, 1-day prediction). A comparison
arXiv:2602.18521v1 Announce Type: new Abstract: Continuous stress forecasting could potentially contribute to lifestyle interventions. This paper presents a novel, explainable, and individualized approach for stress prediction using physiological data from consumer-grade smartwatches. We develop a time series forecasting model that leverages multivariate features, including heart rate variability, activity patterns, and sleep metrics, to predict stress levels across 16 temporal horizons (History window: 3, 5, 7, 9 days; forecasting window: 1, 3, 5, 7 days). Our evaluation involves 16 participants monitored for 10-15 weeks. We evaluate our approach across 16 participants, comparing against state-of-the-art time series models (Informer, TimesNet, PatchTST) and traditional baselines (CNN, LSTM, CNN-LSTM) across multiple temporal horizons. Our model achieved performance with an MSE of 0.053, MAE of 0.190, and RMSE of 0.226 in optimal settings (5-day input, 1-day prediction). A comparison with the baseline models shows that our model outperforms TimesNet, PatchTST, CNN-LSTM, LSTM, and CNN under all conditions, representing improvements of 36.9%, 25.5%, and 21.5% over the best baseline. According to the explanability analysis, sleep metrics are the most dominant and consistent stress predictors (importance: 1.1, consistency: 0.9-1.0), while activity features exhibit high inter-participant variability (0.1-0.2). Most notably, the model captures individual-specific patterns where identical features can have opposing effects across users, validating its personalization capabilities. These findings establish that consumer wearables, combined with adaptive and interpretable deep learning, can deliver relevant stress assessment adapted to individual physiological responses, providing a foundation for scalable, continuous, explainable mental health monitoring in real-world settings.
Executive Summary
This article introduces AdaptStress, a novel approach to stress prediction using multivariate physiological signals from consumer-grade smartwatches. The model leverages time series forecasting and deep learning to predict stress levels across 16 temporal horizons. Evaluating 16 participants over 10-15 weeks, the authors demonstrate that AdaptStress outperforms state-of-the-art time series models and traditional baselines, achieving an MSE of 0.053, MAE of 0.190, and RMSE of 0.226. The model's personalization capabilities are validated through individual-specific patterns and explanability analysis, highlighting the importance of sleep metrics in stress prediction. The findings contribute to scalable, continuous, and explainable mental health monitoring in real-world settings, using consumer wearables and adaptive deep learning.
Key Points
- ▸ AdaptStress is a novel approach to stress prediction using multivariate physiological signals from consumer-grade smartwatches.
- ▸ The model leverages time series forecasting and deep learning to predict stress levels across 16 temporal horizons.
- ▸ AdaptStress outperforms state-of-the-art time series models and traditional baselines in stress prediction.
Merits
Strength in Explainability
The model's ability to provide individual-specific patterns and explanability analysis highlights its personalization capabilities and contributes to scalable, continuous, and explainable mental health monitoring.
Advancements in Time Series Forecasting
The use of multivariate features and deep learning in time series forecasting demonstrates significant improvements in stress prediction accuracy compared to state-of-the-art models.
Demerits
Limited Generalizability
The study's results are based on a relatively small sample size of 16 participants, which may limit the model's generalizability to larger populations.
Dependence on Consumer Wearables
The model's performance is heavily reliant on the quality and accuracy of data from consumer-grade smartwatches, which may introduce biases and limitations.
Expert Commentary
The article presents a significant contribution to the field of predictive analytics in mental health. The use of multivariate features and deep learning in time series forecasting demonstrates a novel approach to stress prediction. However, the study's limitations, such as the small sample size and dependence on consumer wearables, should be addressed in future research. Furthermore, the article highlights the potential of wearable technology in improving mental health outcomes, but its integration and scalability in real-world settings require careful consideration. Overall, the study's findings have practical and policy implications, and further research is needed to explore its potential applications.
Recommendations
- ✓ Future studies should aim to replicate the findings with larger sample sizes and diverse populations to improve generalizability.
- ✓ Developers and policymakers should consider the integration of AdaptStress into wearable devices and mobile applications for real-time stress monitoring and personalized interventions.
