Explainable Artificial Intelligence Models for Accurate Physical Activity Prediction Using Wearable Device Data

Byunggul Lim; Sang-jun Park; Jun-Hyun Bae; Gabrielli T. de Mello; Chad D.Rethorst

doi:10.1016/j.jsams.2026.03.022

April 1, 2026Journal of science and medicine in sportOpen Access

Explainable Artificial Intelligence Models for Accurate Physical Activity Prediction Using Wearable Device Data

View Full Paper

Ask AI

Bookmark

Share

Key Points

The study aims to enhance physical activity prediction accuracy and interpretability using explainable AI models applied to wearable device data.
Conducted a cross-sectional secondary data analysis using publicly available datasets from Apple Watch and Fitbit.
Trained machine learning models (Decision Tree, Random Forest) and deep learning models (GRU, LSTM, CNN-LSTM) to classify physical activity intensity levels.
Assessed model interpretability with Shapley Additive Explanations (SHAP).
Evaluated model performance based on accuracy, precision, recall, and F1-score.
The CNN-LSTM model achieved the highest performance with F1 scores of 82.5 for Apple Watch and 82.3 for Fitbit.
SHAP analysis indicated Apple Watch models emphasized heart rate features during vigorous activity, while Fitbit models focused on step metrics across intensity levels.

No abstract yet

We only have this paper's title and publication details.

Read it at the source

Implication

Cross-sectional analysis explores physical activity predictions using explainable AI in wearable device data, suggesting improvements for health monitoring systems.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Lim et al. (2026) studied this question.

synapsesocial.com/papers/69dc87ea3afacbeac03ea05b https://doi.org/10.1016/j.jsams.2026.03.022

View Full Paper

Ask AI

Bookmark

Share

Explainable Artificial Intelligence Models for Accurate Physical Activity Prediction Using Wearable Device Data

Key Points

No abstract yet

Implication

Discussion

Cite This Study

Explainable Artificial Intelligence Models for Accurate Physical Activity Prediction Using Wearable Device Data

Key Points

No abstract yet

Implication

Discussion

Cite This Study

Keep reading

Keep reading