Design and implementation of a fuzzy intelligent system to estimate the photoplethysmogram systolic features by Persian medicine pulsology

Document Type : Original Article

Authors

1 Department of Mathematics, Payame Noor University

2 Department of Traditional Medicine, School of Persian Medicine, Tehran University of Medical Sciences

3 Payame Noor University, PO Box 3697-19395, Tehran, Iran

4 Ph.D. Student in Applied Mathematics, Payame Noor University

Abstract

Background: In Persian medicine (PM), pulsology is one of the most important ways of knowing the internal states of the body and diagnosing diseases. On the other hand, the photoplethysmogram (PPG) signal provides significant information about heart function for specialists. In this design, fuzzy systems have been used to correlate features from both PPG signal and PM pulsology. Such a system may help pave the path towards the worldwide accepted paradigm of integrative medicine.
Methods: Using the information of 64 individuals, a fuzzy system was designed by MATLAB software. First, the information related to age and pulse parameters including frequency, strength, speed, length, and width were acquired by a PM specialist via traditional pulse examination and were considered as Input variables. Subsequently, their PPG curve was also recorded by a PO80 beurer pulse oximeter. Afterward, variables of slope, height, and time of systolic upstroke curve at points of 25%, 50%, 75%, and 100% amplitude of the PPG curve were calculated and were considered as output variables.
Results: A system with 236 rules was designed and the error of this system was less than 0.05, which is considered an acceptable estimate.
Discussion and conclusion: This designed pilot system estimates the PPG systolic features using PM pulsology with an error of less than 0.05, which can be used to help improve the clinical skills of PM students and practitioners, and can establish the relationship between PM and common medicine.

Keywords

Main Subjects

Health Management & Information Science
Volume 10, Issue 4
October 2023
Pages 226-234

Files

Share

How to cite

Statistics