Temporal Cardiovascular Dynamics for Improved PPG-Based Heart Rate Estimation

2025-10-31

AI Summary

Photoplethysmography (PPG)-based heart rate (HR) estimation suffers from low accuracy in real-world scenarios due to nonlinear and chaotic cardiovascular dynamics. Method: This paper proposes a mutual information-driven temporal dynamical modeling framework integrated with deep learning for HR estimation. Relying solely on single-channel PPG signals—without requiring multimodal sensing or complex post-processing—the method quantifies temporal complexity of HR dynamics via mutual information and constructs chaos-aware feature representations, which are embedded into a lightweight deep architecture for end-to-end estimation. Results: Evaluated on four publicly available datasets covering diverse real-world conditions—including physical exercise and daily activities—the proposed approach reduces mean absolute HR estimation error by up to 40% compared to conventional filtering methods and state-of-the-art machine learning models. This significantly enhances the robustness and practicality of cardiovascular health monitoring in wearable devices.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Hardware-aware ML
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The oscillations of the human heart rate are inherently complex and non-linear -- they are best described by mathematical chaos, and they present a challenge when applied to the practical domain of cardiovascular health monitoring in everyday life. In this work, we study the non-linear chaotic behavior of heart rate through mutual information and introduce a novel approach for enhancing heart rate estimation in real-life conditions. Our proposed approach not only explains and handles the non-linear temporal complexity from a mathematical perspective but also improves the deep learning solutions when combined with them. We validate our proposed method on four established datasets from real-life scenarios and compare its performance with existing algorithms thoroughly with extensive ablation experiments. Our results demonstrate a substantial improvement, up to 40%, of the proposed approach in estimating heart rate compared to traditional methods and existing machine-learning techniques while reducing the reliance on multiple sensing modalities and eliminating the need for post-processing steps.

Problem

• Modeling non-linear chaotic heart rate dynamics
• Improving PPG-based heart rate estimation accuracy
• Reducing reliance on multiple sensing modalities

Innovation

• Modeling heart rate with non-linear chaotic dynamics
• Combining mathematical analysis with deep learning
• Reducing sensor dependency and post-processing steps