<![CDATA[Abstract— Photoplethysmography (PPG) is a non-invasive optical technique for cardiovascular health monitoring, such as blood pressure estimation and arterial stiffness analysis. However, detecting fiducial points in PPG signals such as the onset, systolic peak, dicrotic notch, and diastolic peak is often hindered by noise, baseline wander, and physiological variability. Although various methods have been proposed, such as time-frequency domain analysis and machine learning algorithms, these approaches still have limitations, including high computational complexity and susceptibility to noise. This study proposes a gradient-based analysis approach to improve the accuracy of fiducial point detection in PPG signals. The gradient method is used to detect local maxima and minima in the PPG signal. By incorporating validation and correction modules based on temporal order and amplitude ratios, the approach achieves 100% detection accuracy after initial error correction (initial error rate: 58% for the dicrotic notch). The results demonstrate that this method effectively identifies all fiducial points (onset, systolic peak, dicrotic notch, diastolic peak) in 50 out of 50 datasets, with robust performance against noise and physiological variability. This study confirms that the gradient-based method is suitable for cost-efficient, portable diagnostic applications. ]]>
Copyrights © 2025
