Parameter identification and identifiability analysis for patient-induced effort in respiratory mechanics models

Johnston Lee Teong Jeen, Chiew Yeong Shiong, Ganesaramachandran Arunachalam

Research output: Chapter in Book/Report/Conference proceedingConference PaperOther

Abstract

A growing concern in the field of mechanical ventilation (MV) treatment is the lack of optimal patient-specific ventilator setting and automation for spontaneously breathing patients. Model-based respiratory mechanics have been introduced as a non-invasive method of performing respiratory mechanics estimation. In volume-control ventilation (VC), patient’s spontaneous effort causes anomalies in the airway pressure reading, which cause alterations in the airway pressure delivered. As a result, MV patient’s respiratory mechanics cannot be determined for treatment purposes. To remedy this, a ‘polynomial model’ was investigated and the results show a practically nonidentifiable model. In this study, a ‘sine-wave model’ where a sinusoidal function added to the single compartment model to describe and capture the patient effort is presented. Monte-Carlo Analysis and identifiability analysis was carried out to determine the performance and stability of the sine-wave model. The results of this study indicate that the sine-wave model fits pressure waveforms with patient effort well but it is practically nonidentifiable in some cases. Misidentification of respiratory mechanics arises when critical characteristics of the pressure waveform are unexpressed. Future research should be focused on models that are not extensions of the single compartment model.

Original languageEnglish
Title of host publicationProceedings of ICIBEL 2019
EditorsFatimah Ibrahim, Juliana Usman, Mohd Yazed Ahmad, Norhamizan Hamzah
Place of PublicationCham Switzerland
PublisherSpringer
Pages3-13
Number of pages11
ISBN (Electronic)9783030650926
ISBN (Print)9783030650919
DOIs
Publication statusPublished - 2021
EventInternational Conference for Innovation in Biomedical Engineering and Life Sciences 2020 - Kuala Lumpur, Malaysia
Duration: 6 Dec 20197 Dec 2019
Conference number: 3es
https://link.springer.com/book/10.1007/978-3-030-65092-6 (Proceedings)

Publication series

NameIFMBE Proceedings
PublisherSpringer
Volume81
ISSN (Print)1680-0737
ISSN (Electronic)1433-9277

Conference

ConferenceInternational Conference for Innovation in Biomedical Engineering and Life Sciences 2020
Abbreviated titleICIBEL 2020
Country/TerritoryMalaysia
CityKuala Lumpur
Period6/12/197/12/19
Internet address

Keywords

  • Identifiability analysis
  • Mechanical ventilation
  • Parameter identification
  • Respiratory mechanics models

Cite this