PPG signal processing for estimation of blood pressure using anfis method – Application for medical robot

Abstract

Photoplethysmography (PPG) is a noninvasive method for detecting vital indicators such as blood pressure, heart rate, and the saturation of peripheral oxygen (SpO2) (BP). To make it simple for doctors to keep an eye on patients' health, a PPG processing system was proposed to estimate blood pressure (BP). This system is applied to Medical Robot. Motion artifacts were removed by using Min Max Scaler Normalization and 20 Hz cut-off frequency Low Pass filter. As a result, data accuracy and quality have increased noticeably. We suggested an AI technical system named the Adaptive Neuro-Fuzzy Inference System (ANFIS) based on the ANN algorithm to assist in estimating critical indicators like SBP and DBP. Python and MATLAB are the two primary languages and tools for processing the incoming data to generate the most accurate outputFifteen records (15 patients) were sensed and used for training and testing purposes. The Root Mean Squared Error (RMSE) of the testing data were 0.23 and 0.11 respectively for Diastolic Blood Pressure (DBP) and Systolic Blood Pressure(SBP). The conventional and specialized system for BP estimation was statistically proven. Building an IOT system that records vital data into a database and then displays it on mobile applications, PC screens, etc. is a solution to letting slimmers monitor patients more quickly, more efficiently, and more easily