Nowadays, Blood Pressure (BP) is an established major risk factor to determine cardiovascular accidents. Current BP monitors are not able to deal with noisy situations such as those present in stress tests. The aim of this study is to develop a system able to measure the BP even under these conditions. A device and an algorithm based on an Artificial Neural Network (ANN) are proposed as a feasible solution for BP measurement. Different ANN structures were trained to optimize the recognition of the Korotkoff sounds and the best implemented in the final system. The system generates measurements similar to those of the physician, yet the differences increase with high deflation rates or when the staff lecture is not reliable. This work demonstrates the feasibility of an ANN application to improve BP measurement in situations in which the most reliable method known at present is conducted manually. © Springer-Verlag Berlin Heidelberg 2013.
CITATION STYLE
Salazar Mendiola, J. L., Vargas Luna, J. L., González Guerra, J. L., & Cortés Ramírez, J. A. (2013). Application of a Neural Network to Improve the Automatic Measurement of Blood Pressure. In Communications in Computer and Information Science (Vol. 383 CCIS, pp. 263–272). Springer Verlag. https://doi.org/10.1007/978-3-642-41013-0_27
Mendeley helps you to discover research relevant for your work.