Artery Research

Volume 25, Issue Supplement 1, December 2019, Pages S16 - S18

2.7 Machine Learning on Central Hemodynamic Quantities Using Noninvasive Measurements: How Far Can We Go?

Authors
Vasiliki Bikia*, Stamatia Pagoulatou, Nikolaos Stergiopulos
Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland
*Corresponding author. Email: vasiliki.bikia@epfl.ch
Corresponding Author
Vasiliki Bikia
Available Online 15 February 2020.
DOI
10.2991/artres.k.191224.012How to use a DOI?
Abstract

Background: Transforming peripheral noninvasive measurements to obtain central hemodynamic quantities, such as cardiac output (CO) and central systolic blood pressure (cSBP), is a highly emerging field [1,2]. However, no holistic investigation has been performed to assess the amount of information contained in each peripheral measurement for the prediction of central values. This can be attributed to the inherent difficulty of creating a complete and accurate database; mainly due to the invasive nature of the gold standard techniques [3,4].

Methods: To meet this need, we exploit synthetic data from a previously validated cardiovascular model (CVm) [5]. Our study relies on peripheral quantities including brachial pressure, heart rate (HR), and pulse wave velocity (PWV) simulated by the CVm. A Random Forest model was trained using 2744 synthetic instances and, subsequently, was tested against a subset of 800. Correlations and feature importances of the input parameters were reported (Figure 1).

Results: Our results demonstrated that precise estimates of CO and cSBP were yielded with an RMSE of 0.39 L/min and 1.39 mmHg, respectively (Figures 2 and 3). Low biases were observed, namely 0.03 ± 0.39 L/min for CO and −0.08 ± 1.39 mmHg for cSBP. PWV, HR, and brachial pulse pressure were found to be the most correlated features with CO, whereas brachial SBP was plausibly shown to be the significant determinant of cSBP for our model (Figures 4 and 5).

Conclusion: These findings pave the way for better devising central hemodynamics’ predictions. In the future, our ultimate goal is to examine the sensitivity of cardiac parameters estimation (i.e., elastance) to noninvasive peripheral measurements.

Figure 1

Correlation matrix.

Figure 2

Scatterplot (A) and Bland-Altman plot (B) between predicted and reference CO values.

Figure 3

Scatterplot (A) and Bland-Altman plot (B) between predicted and reference CO values.

Figure 4

Feature importances for CO prediction.

Figure 5

Feature importances for cSBP.

Copyright
© 2019 Association for Research into Arterial Structure and Physiology. Publishing services by Atlantis Press International B.V.
Open Access
This is an open access article distributed under the CC BY-NC 4.0 license (http://creativecommons.org/licenses/by-nc/4.0/).

Journal
Artery Research
Volume-Issue
25 - Supplement 1
Pages
S16 - S18
Publication Date
2020/02/15
ISSN (Online)
1876-4401
ISSN (Print)
1872-9312
DOI
10.2991/artres.k.191224.012How to use a DOI?
Copyright
© 2019 Association for Research into Arterial Structure and Physiology. Publishing services by Atlantis Press International B.V.
Open Access
This is an open access article distributed under the CC BY-NC 4.0 license (http://creativecommons.org/licenses/by-nc/4.0/).

Cite this article

TY  - JOUR
AU  - Vasiliki Bikia
AU  - Stamatia Pagoulatou
AU  - Nikolaos Stergiopulos
PY  - 2020
DA  - 2020/02/15
TI  - 2.7 Machine Learning on Central Hemodynamic Quantities Using Noninvasive Measurements: How Far Can We Go?
JO  - Artery Research
SP  - S16
EP  - S18
VL  - 25
IS  - Supplement 1
SN  - 1876-4401
UR  - https://doi.org/10.2991/artres.k.191224.012
DO  - 10.2991/artres.k.191224.012
ID  - Bikia2020
ER  -