From the seashore to the stratosphere: A step closer to continuous lung monitoring using wearable ultrasound

Lung sliding quantification

Master thesis (2022)

Authors

J.M. Visser Mechanical Engineering

Contributors

D. van Westerloo Leiden University Medical Center (supervisor 1)
J.E. Lopez Matta Leiden University Medical Center (supervisor 1)
H. Pham Philips Research (supervisor 2)
E. de Jonge Leiden University Medical Center (supervisor 2)
C. Elzo Kraemer Leiden University Medical Center (supervisor 1)
R. Dekker Electronic Components, Technology and Materials - EEMCS (supervisor 2)
Faculty
Mechanical Engineering
Copyright: © 2022 Jeffrey Visser
More Info
expand_more

Published Date

01-03-2022

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Mechanical Engineering

Abstract

Background: The prevention of ventilator-induced lung injury (VILI) is an important topic in criticalcare. Lung overdistention is a great attributor to VILI. Regional evaluation of the pulmonary mechanicsis a relatively new method and is currently being studied using electrical impedance tomography (EIT)and computed tomography (CT). However, ultrasonography may be an effective alternative tomeasure lung overdistention by quantifying lung sliding.
Objective: The primary aim of this study was to explore if lung sliding can be quantified using in-housemade speckle tracking algorithms. Second, the usability of lung sliding quantification to detect lungoverdistention was investigated.
Methods: Two speckle tracking algorithms were built and validated to quantify lung sliding. Adultpatients admitted to the ICU were prospectively examined using ultrasound. Ultrasound images wereanalyzed offline using the in-house developed Fourier-Based Speckle Tracking algorithm (FBST) andthe Intensity-Based Speckle Tracking algorithm (IBST). The performance of the best algorithm wasvalidated against the manual evaluation of lung sliding by experts and validated against lungcompliance. Besides, the performance of the best algorithm was investigated by reproducibilitytesting. Finally, we tested the algorithm for its ability to detect lung overdistention.
Results: The FBST algorithm could not reliably quantify lung sliding and was not further evaluated.However, the IBST algorithm showed an overall success rate of 88%. Moreover, the IBST algorithmdifferentiated normal and moderate lung sliding with a mean IBST score of 12.1 and 29, respectively(p<0.001). A positive correlation was measured between IBST score and lung compliance (β 0.15, SE0.025, 95% CI 0.101-0.208, p <0.001). The reproducibility test calculated a relative standard error of2.7. Finally, a difference was found in IBST scores between patients with a PEEP value <12 and ≥12with a mean ± SD of 20.9 ± 7 and 12.6 ± 6, p < 0.05, respectively.
Conclusion: The IBST algorithm enables a robust lung sliding quantification. The IBST scores correlatewell with expert opinion and lung compliance, but the reliability needs improvement. Our results showthat detection of lung overdistention using ultrasound is feasible. Further clinical studies should assessif ultrasonography can measure lung overdistention.