A Laplace diffusion-based transformer model for heart rate forecasting within daily activity context

2025-08-20

AI Summary

To address prediction bias in remote heart rate monitoring from wearable devices caused by neglecting activity context, this paper proposes an activity-aware temporal modeling approach. We integrate a Laplacian diffusion mechanism into the Transformer architecture to capture the nonlinear dynamic coupling between heart rate and daily activities. Additionally, we design activity-specific embeddings and a context-aware attention module to precisely model activity-driven heart rate variations. The model jointly processes heart rate time series and multi-class activity labels via dedicated encoders that co-learn physiological and behavioral features. Evaluated on a real-world patient cohort, our method achieves a 43% reduction in mean absolute error and an R² of 0.97 compared to state-of-the-art baselines, demonstrating substantial improvements in both predictive accuracy and clinical interpretability.

Technology Category

• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Hardware-aware ML

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

With the advent of wearable Internet of Things (IoT) devices, remote patient monitoring (RPM) emerged as a promising solution for managing heart failure. However, the heart rate can fluctuate significantly due to various factors, and without correlating it to the patient's actual physical activity, it becomes difficult to assess whether changes are significant. Although Artificial Intelligence (AI) models may enhance the accuracy and contextual understanding of remote heart rate monitoring, the integration of activity data is still rarely addressed. In this paper, we propose a Transformer model combined with a Laplace diffusion technique to model heart rate fluctuations driven by physical activity of the patient. Unlike prior models that treat activity as secondary, our approach conditions the entire modeling process on activity context using specialized embeddings and attention mechanisms to prioritize activity specific historical patents. The model captures both long-term patterns and activity-specific heart rate dynamics by incorporating contextualized embeddings and dedicated encoder. The Transformer model was validated on a real-world dataset collected from 29 patients over a 4-month period. Experimental results show that our model outperforms current state-of-the-art methods, achieving a 43% reduction in mean absolute error compared to the considered baseline models. Moreover, the coefficient of determination R2 is 0.97 indicating the model predicted heart rate is in strong agreement with actual heart rate values. These findings suggest that the proposed model is a practical and effective tool for supporting both healthcare providers and remote patient monitoring systems.

Problem

• Forecasting heart rate fluctuations during daily physical activities
• Integrating activity context into heart rate monitoring models
• Improving accuracy of remote patient monitoring using AI

Innovation

• Transformer model with Laplace diffusion technique
• Activity context conditioning with specialized embeddings
• Contextualized embeddings and dedicated encoder integration