Research Article
Machine Learning Approach for Blood Pressure Measurement Using Bio-Impedance
@INPROCEEDINGS{10.4108/eai.17-11-2023.2342840, author={Kanthalakshmi S and Bhavishyashri C S}, title={Machine Learning Approach for Blood Pressure Measurement Using Bio-Impedance}, proceedings={Proceedings of the First International Conference on Science, Engineering and Technology Practices for Sustainable Development, ICSETPSD 2023, 17th-18th November 2023, Coimbatore, Tamilnadu, India}, publisher={EAI}, proceedings_a={ICSETPSD}, year={2024}, month={1}, keywords={bio-impedance xg boost algorithm impedance plethysmography}, doi={10.4108/eai.17-11-2023.2342840} }- Kanthalakshmi S
Bhavishyashri C S
Year: 2024
Machine Learning Approach for Blood Pressure Measurement Using Bio-Impedance
ICSETPSD
EAI
DOI: 10.4108/eai.17-11-2023.2342840
Abstract
Continuous monitoring of flow of blood in the blood vessel is of paramount importance in heart related healthcare, particularly in the management of hypertension. Impedance plethysmography (IPG) is a non-invasive diagnostic method used for signal measurement and assessing the condition of a patient’s arteries, specifically the carotid artery. In order to achieve accurate IPG-based carotid pulse detection for heart related diagnostic applications, this paper presents a set of optimized measurement parameters to effectively capture the pulsations originating from the carotid artery. The analysis in this study explores the influence of factors such as the excitation current frequency, the electrode crosssectional area, electrode arrangements, and the physiological location of the carotid arteries on the resolution of IPG measurements. This research contributes to the efficient measurement of bioimpedance, facilitating the accurate determination of Blood Pressure (BP) values. In this study, electrodes are placed over carotid artery to measure bio-impedance. The obtained impedance value is collected as database and a model is created using XGBoost regression algorithm. This model is trained and tested for accuracy measures by estimating RMSE and ME values with standard criteria.
