cs.LGFeb 19, 2026

AdaptStress: Online Adaptive Learning for Interpretable and Personalized Stress Prediction Using Multivariate and Sparse Physiological Signals

Authors: Xueyi Wang, Claudine J. C. Lamoth, Elisabeth Wilhelm

Organizations: Engineering and Technology institute Groningen (ENTEG), University of Groningen, Groningen, 9742AG, the Netherlands · Department of Human Movement Sciences, University Medical Center Groningen, Groningen, Netherlands

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

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.

Figures & tables

Fig. 1: Individual feature analysis of physiological data distribution.
Feature NameAbbrev.RangeUnit
Total KilocaloriesTK381-4726kcal
Total StepsTS46-31374steps
Total Distance MetersTD31-80528meters
Highly Active SecondsHA0-25768seconds
Active SecondsACS0-21357seconds
Moderate Intensity MinutesMI0-961minutes
TABLE I: Feature List with Abbreviations, ranges, and units
Fig. 2: Missing and anomaly numbers and rates for all features
Fig. 3: Domain shifts illustrated by UMAP analysis for features of each participant
Fig. 4: Diagram of AdaptStress model used in this work
Fig. 5: Comprehensive performance comparison across all history window sizes (3, 5, 7, 9 days) and prediction horizons (1, 3, 5, 7 days) for seven different forecasting models: our proposed approach, state-of-the-art time series models (Informer, TimesNet, PatchTST), and traditional deep learning baselines (CNN-LSTM, LSTM, CNN) by three evaluation metrics (Mean Absolute Error, Mean Squared Error, Root Mean Squared Error).
MetricModelInput = 3 daysInput = 5 daysInput = 7 daysInput = 9 days
1357135713571357
MAELSTM0.2500.2600.2810.2680.2550.2720.2800.2830.2680.2690.2790.2730.2570.2690.2710.265
CNN0.2380.2470.2560.2660.2440.2520.2630.2740.2540.2600.2610.2680.2530.2630.2610.271
CNN-LSTM0.2380.2600.2550.2610.2470.2500.2540.2680.2570.2560.2610.2600.2510.2640.2540.266
Informer0.2370.2390.2460.2480.2440.2510.2450.2540.2450.2500.2520.2490.2480.2470.2500.255
PatchTST0.2520.2570.2590.2660.2540.2670.2730.2730.2610.2710.2820.2680.2540.2700.2840.282
TABLE II: Comprehensive performance comparison across different metrics, input, models, and prediction window sizes
Fig. 6: Comparison of ground truth and predicted stress values for eight representative participants. Solid blue lines indicate actual stress scores, with yellow dotted lines showing validation predictions and red lines showing test predictions. Shaded regions represent prediction confidence intervals. Bottom bars indicate trend direction accuracy (green: correct, blue: incorrect).
Fig. 7: Radar plot comparison of prediction errors (MAE, MSE, RMSE) across 16 participants for all seven models (H5-P1). Smaller areas indicate better performance.
Fig. 8: Feature importance, consistency, and individual-specific patterns in stress prediction revealed through SHAP analysis.

Explore similar work

CardsList
Jun 23, 2026

Retrieval-Augmented Personalization with Foundation Models for Wearable Stress Detection

Personalization in wearable-based stress detection remains challenging due to substantial inter-individual variability in physiological and behavioral responses. While traditional approaches rely on user-specific fine-tuning or costly self-supervised pre-training on large datasets, we propose a lightweight alternative based on retrieval-augmented personalization. Our method leverages frozen, out-of-domain foundation models to retrieve similar patterns from a target user's history and encode them into a compact personalized embedding that modulates representations extracted by a lightweight transformer network. We evaluate our approach on the WESAD stress detection dataset with N=15 users, comprising wrist-worn physiological (EDA, BVP, temperature) and activity (accelerometer) signals, and report gains of +3.92% in accuracy and +4.76% in macro F1-score over a non-personalized transformer baseline, approaching supervised fine-tuning performance without requiring any labeled user data. We further show that temporal retrieval, where only prior user samples are available, achieves performance close to full intra-user retrieval, demonstrating robustness to limited user history. Finally, we explore personalization in a cross-dataset retrieval setting, leveraging embeddings from the K-Emocon dataset to personalize representations for stress detection on the WESAD dataset.

Louis Simon, Mohamed Chetouani

Institut des Systèmes Intelligents et de Robotique Sorbonne University Paris, France

Wearable Health Monitoring
AskPDF
Apr 16, 2026

Personalized and Context-Aware Transformer Models for Predicting Post-Intervention Physiological Responses from Wearable Sensor Data

Consumer wearables enable continuous measurement of physiological data related to stress and recovery, but turning these streams into actionable, personalized stress-management recommendations remains a challenge. In practice, users often do not know how a given intervention, defined as an activity intended to reduce stress, will affect heart rate (HR), heart rate variability (HRV), or inter-beat intervals (BBI) over the next 15 to 120 minutes. We present a framework that predicts post-intervention trajectories and the direction of change for these physiological indicators across time windows. Our methodology combines a Transformer model for multi-horizon trajectories of percent change relative to a pre-intervention baseline, direction-of-change calls (positive, negative, or neutral) at each horizon, and an empirical study using wearable sensor data overlaid with user-tagged events and interventions. This proof of concept shows that personalized post-intervention prediction is feasible. We encourage future integration into stress-management tools for personalized intervention recommendations tailored to each person's day following further validation in larger studies and, where applicable, appropriate regulatory review.

Esther Brown, Victoria Dean, Finale Doshi-Velez

Harvard University, Cambridge, MA · Princeton University, Princeton, NJ

Multivariate Time Series ForecastingWearable Health Monitoring
AskPDF
Jul 23, 2026

Explainable graph attention network for stress recognition (StressGAT) via differential action units

Stress is a dynamic process characterized by significant individual variability in facial expression. Traditional architectures, such as Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs), often overlook person-specific baselines or lack the representational capacity to model the non-linear temporal progression of distress due to sequential bottlenecks and rigid grid-based constraints. Furthermore, many deep learning models lack the interpretability required for clinical deployment. This study introduces StressGAT, a Graph Attention Network that leverages the relational inductive bias of graph modeling to capture complex facial dynamics that indicate acute stress. By using Differential Action Units, the framework normalizes individual responses relative to neutral baselines to achieve personalized recognition. The proposed model achieves 88.62% accuracy on a diverse stress-induction cohort (58 participants) using a subject-independent, Leave-One-Subject-Out (LOSO) cross-validation protocol. Beyond predictive accuracy, the architecture integrates a Multiple Instance Learning (MIL) attention mechanism to identify peak stress intervals and reveal distinct expressivity phenotypes. By simultaneously optimizing for accuracy and interpretability, this framework provides a robust, explainable solution for personalized affective monitoring.

Thomas Kassiotis, Stefanos Gkikas, Nikolaos Smyrnis +1

Department of Electronic Engineering Hellenic Mediterranean University Chania, Greece · Honda Research Institute Japan Wako City, Japan · University Mental Health, Neurosciences and Precision Medicine, Research Institute “COSTAS STEFANIS” 2nd Department of Psychiatry, Medical School, National and Kapodistrian University of Athens Athens, Greece +1

Graph Attention NetworksFacial Expression RecognitionAffective Computing
AskPDF