ARTERIAL STIFFNESS MODELING USING WINDKESSEL MODEL FOR EARLY DETECTION OF CARDIOVASCULAR DISEASE

• Main Author: Masita Dewi, Ervin
• Format: Dissertations
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/63055
• Institution: Institut Teknologi Bandung

ABSTRACT ARTERIAL STIFFNESS MODELING USING WINDKESSEL MODEL FOR EARLY DETECTION OF CARDIOVASCULAR DISEASE By ERVIN MASITA DEWI NIM: 33215005 (Doctoral Program in Electrical Engineering and Informatics) Cardiovascular disease is the main cause of death in the world which is generally caused by aging and lifestyle factors. Current diagnostic methods detect disease at an advanced stage, ie at the onset of symptoms, not at an early stage. Biomedical Engineering Expertise Group, School of Electrical and Informatics Engineering (STEI), Bandung Institute of Technology (ITB) developed a device to detect cardiovascular disease. The tool, named NIVA or Non-Invasive Vascular Analyzer, is a non-invasive device that uses Photoplethysmography (PPG) sensors and blood pressure sensors to evaluate blood vessels, one of which is predicting reduced elasticity of blood vessels. The parameters of the NIVA measurement results to measure the level of vessel elasticity are the Stiffness Index (SI) and Augmentation Index (AI). To improve the performance of NIVA, a model that can estimate the waveform is made to describe the condition of blood vessel stiffness. The model is based on PPG sensor measurement data from NIVA on the fingertips and toes that have not been utilized. The model chosen is the Windkessel model which has several types, ranging from single-segment Windkessel to multi-segment. All of them are used to see the pattern of blood flow that occurs in the body. The advantage of this research is the addition of leg segments in the developed model. The addition of the leg segment results in a reflected wave, from this reflected wave the level of stiffness of the blood vessels is analyzed. Apart from adding leg segments, this modeling is equipped with measurement data for heart patients, so that the simulation results from the developed model have been validated against the results of measuring the stiffness of blood vessels from heart patients. Therefore, this developed model can be used for early detection of cardiovascular disease. The method used in this study compared arterial stiffness index through PPG signal analysis between healthy people and cardiac patients who had undergone catheterization. The data used is the result of the NIVA measurement in the form of a PPG signal measured from the tips of the fingers and toes. Data collection was carried out on 10 healthy people and 10 heart patients at the Diagnostic and Cardiac Center of Hasan Sadikin Hospital, Bandung. With the Anova test result of p < 0.05, it means that there is a significant difference between the data of healthy people and the data of cardiac catheterization patients, so that the measurement data obtained can be used as a reference for model making. This research has succeeded in developing a model of vascular stiffness based on the multi- segment Windkessel model with 3 hand segments, 3 leg segments, and 1 aortic segment. By using this model, it is possible to estimate the waveform of the PPG signal at the fingertips and the tip of the left toe according to the level of stiffness of the blood vessels. The blood vessel segments that have the most influence on the level of stiffness are the subclavian artery, abdominal aorta, and iliac segments. From the developed model, it is found that there is a relationship between the capacitance value of the Windkessel model and the vessel stiffness parameter. Changes in the value of C greatly affect the waveform. The waveform due to changes in the value of C, especially in segments 3, 4 and 5 illustrates the condition of blood vessel stiffness. In the physiological condition of cardiovascular patients with a C value that is reduced by 20% from the normal value, the SI value is 608 cm/s for the hands and 345 cm/s for the feet, where the normal value for healthy people is the C SI value is 365 cm/s for the hands. and 288 cm/s. By setting the value of C on the model, we can see the PPG waveform of the fingertips and toes. A 20% decrease in the capacitance value results in an increase in SI of 66% and will increase the risk of cardiovascular disease by 80%. Keywords: Photoplethysmography, Stiffness Index, Augmentation Index, Windkessel.