Articolul precedent |
Articolul urmator |
229 2 |
Ultima descărcare din IBN: 2023-10-30 19:29 |
SM ISO690:2012 KUCUR, Murad, BILGIC, Bogac. Estimation of brachial blood pressure for individuals aged 25 years with multiple linear regression. In: Conference on Applied and Industrial Mathematics: CAIM 2021, 17-18 septembrie 2021, Iași, România. Chișinău, Republica Moldova: Casa Editorial-Poligrafică „Bons Offices”, 2021, Ediţia a 28-a, p. 29. |
EXPORT metadate: Google Scholar Crossref CERIF DataCite Dublin Core |
Conference on Applied and Industrial Mathematics Ediţia a 28-a, 2021 |
|||||
Conferința "Conference on Applied and Industrial Mathematics" Iași, România, Romania, 17-18 septembrie 2021 | |||||
|
|||||
Pag. 29-29 | |||||
|
|||||
Descarcă PDF | |||||
Rezumat | |||||
The owrate, velocity and pressure changes are important parameters in the hemodynamic analysis of blood ow in the vessels.The pressure variation during blood ow plays an important role in hemodynamic studies. In CFD studies the pressure, owrate and velocity inlet or outlet boundary conditions are e ective for precise and accurate analysis. Specifying the pressure wave obtained from real patient datas as an inlet or outlet condition during the CFD analysis ensures that the analysis is close to reality. For this reason, blood pressure waves can be obtained using mathematical models, and very close to real blood pressure waves can be estimated with Machine Learning algorithms. In this study, the blood pressure wave was tried to be estimated by taking various characteristics of 25-year-old individuals. The available data was analyzed rst and it was determined which parameters played an active role. Then % 70 of this data was chosen as training data and the rest was reserved as test data. Considering the important parameters, a multiple linear regression model was made and pressure waves were created with the existing parameters thanks to the model. With the test data, it was evaluated how successful these predictions were. Virtual pressure waves required for other studies could be obtained from the model that was used in this study. |
|||||
Cuvinte-cheie blood pressure, machine learning, multiple linear regression |
|||||
|