Vargas Juan M, Boularas Mohamed M, Bahloul Mohamed A, Aridhi Slah, Laleg-Kirati Meriem
Annu Int Conf IEEE Eng Med Biol Soc. 2024 Jul;2024:1-6. doi: 10.1109/EMBC53108.2024.10782510.
In this paper, we propose a data-driven model lever-aging a Limited Penetrable Weighted Visibility Graph (LPWVG) derived from the photoplethysmogram (PPG) waveform for Pulse Wave Velocity (PWV) estimation. Four distinct LPWVGs were constructed employing diverse weighted methods. Subsequently, various features have been computed and extracted from PPG images, including two-dimensional Semi-classical Signal Analysis (SCSA)-based features, frequency-based features, and shape-based features. These features were then input into different machine-learning models. The proposed approach's performance was rigorously evaluated using both in-silico and real PPG pulse wave data. The obtained results provide compelling evidence supporting the feasibility and effectiveness of the proposed method for accurate PWV estimation in biomedical applications.
在本文中，我们提出了一种数据驱动模型，该模型利用从光电容积脉搏波（PPG）波形导出的有限穿透加权可见性图（LPWVG）来估计脉搏波速度（PWV）。采用不同的加权方法构建了四种不同的LPWVG。随后，从PPG图像中计算并提取了各种特征，包括基于二维半经典信号分析（SCSA）的特征、基于频率的特征和基于形状的特征。然后将这些特征输入到不同的机器学习模型中。使用计算机模拟和真实PPG脉搏波数据对所提出方法的性能进行了严格评估。所得结果提供了有力证据，支持了该方法在生物医学应用中准确估计PWV的可行性和有效性。
