Back to search

University of Southampton

Model predictive valve control to assist in tracking a lung pressure profile

Abstract

dc:description.abstract

In the UK, an estimated 88,000 people have a brain tumour, and typically are unaware of its presence until symptoms occur. Currently there is no mass screening available due to limitations in diagnostic techniques. Measuring changes in intracranial pressure could be revolutionary for diagnosing many cerebral pathologies. A strong link between intracranial and inner ear pressure is known to exist. Thus the indirect measurement of intracranial pressure, via tympanic membrane displacement, is a potential low-cost, accessible solution. However, natural physiological pressure fluctuations distort this association. Forced expiration during tympanic membrane displacement has the potential of reducing this distortion. <br/><br/>Lung pressure profiling is a recognised procedure in which a person generates specific lung pressures at specific times. These pressures are measured by using a hand-held breathing apparatus. A lung pressure profile reference is provided for a person to track during forced expiration. The pressures, and changes in pressures, when tracking the reference should result in corresponding changes in intracranial pressure. These intracranial pressure changes would be observed in tympanic membrane displacement. However, a person may not be able to accurately track the reference using the current clinical research breathing apparatus.<br/><br/>This thesis develops a solution to the tracking problem by assisting participants to precisely track pressure profile references, achieved by controlling airflow during forced expiration. This stabilises intrathoracic pressure, significantly reducing the physiological pressure fluctuations. The thesis develops and evaluates the first model to replicate a person's lung pressure profile tracking response. The airflow controller uses a novel model predictive control framework to adjust a model parameter in order to assist the person's tracking response. A clinically-feasible identification approach is then derived. Results with 10 participants confirm that lung pressure profile reference tracking is improved by an average of 22\% compared to the current clinical research approach.

Degree

thesis:*
Name dc:type.qualificationname
Ph.D.
Level dc:type.qualificationlevel
doctoral
Grantor dc:publisher.institution
University of Southampton
Year dc:date.issued
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Thompson, Michael Callum
Advisors dc:contributor.advisor
  • Freeman, Chris
  • Hughes, Ann-Marie
  • O'Brien, Neil Stephen
  • Marchbanks, Robert J.
  • Birch, Tony
ORCID person:identifier.orcid

Rights

Language dc:language
English, eng

Chain of custody

source
Harvested from
University of Southampton
Base URL
eprints.soton.ac.uk/cgi/oai2
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Thompson, Michael Callum. Model predictive valve control to assist in tracking a lung pressure profile. doctoral thesis, University of Southampton, 2025.