Photoplethysmography (PPG) Signal-Based Chronological Age Estimation Algorithm

Chronological age estimation is essential in the medical and forensic world because it is the biological profile of each individual. The purpose of this study is to present a chronological age estimation algorithm based on Photoplethysmography (PPG) signals. The algorithm is divided into four stages: preprocessing, feature extraction, classification, and estimation. The preprocessing stages include low-pass filter, normalization, and signal cutting. In the feature extraction stage, one signal cycle produces 10 features extracted from the PPG signal and 20 features extracted from the APG signal (Second Derivative of PPG Signal). These features are used in the classification stage by applying the Support Vector Machine (SVM) method. The age estimate is obtained by applying a linear kernel to the SVM method. To test the performance of the algorithm, validation is carried out by calculating accuracy and Mean Absolute Percentage Error (MAPE). Based on the validation results, the accuracy value is 91.8% and MAPE is 8.2%. These results indicate that the PPG method has the potential to be used as an alternative method for estimating chronological age.