Related Experiment Video
Updated: Jun 4, 2025
08:22
Author Spotlight: Advancing the Study of Brain-Heart Interplay with a Comprehensive EEGLAB Plugin for Multimodal Signal Analysis
Published on: April 26, 2024
1.7K
A low-cost PPG sensor-based empirical study on healthy aging based on changes in PPG morphology.
Muhammad Saran Khalid1, Ikramah Shahid Quraishi1, Muhammad Wasim Nawaz2
1Electrical engineering department, Information Technology University, Lahore, Pakistan.
Physiological Measurement
|December 20, 2024
Summary
Photoplethysmography (PPG) signals can estimate vascular and chronological age. This low-cost, non-invasive sensor accurately predicts age groups and individual ages, showing promise for healthy aging research.
Area of Science:
- Biomedical Engineering
- Gerontology
- Signal Processing
Background:
- Photoplethysmography (PPG) is a non-invasive optical technique to detect blood volume changes.
- Understanding healthy aging is crucial for public health and personalized medicine.
- Existing methods for age estimation may be invasive or costly.
Purpose of the Study:
- To investigate changes in PPG signal morphology related to healthy aging.
- To develop a low-cost, non-invasive method for estimating vascular age, chronological age, and age group.
- To correlate PPG signal features with physiological aging markers.
Main Methods:
- Collected raw infrared PPG data from 173 healthy South Asian subjects (aged 3-61).
- Extracted 62 features from conditioned PPG signals, reduced to 26 key features using correlation-based ranking.
- Applied machine learning classifiers including logistic regression, random forest, XGBoost, feedforward neural networks, and convolutional neural networks.
Main Results:
- XGBoost achieved 99% accuracy in classifying age groups (binary and three-class).
- Random forest demonstrated a mean absolute error of 6.97 years for vascular/chronological age prediction.
- The study identified significant correlations between PPG features and aging.
Conclusions:
- PPG signals offer a promising, low-cost, non-invasive biomarker for studying healthy aging.
- Machine learning models can effectively utilize PPG features for age estimation.
- This approach has potential applications in health monitoring and personalized aging assessments.
