Use this URL to cite or link to this record in EThOS:
Title: Non-contact monitoring of respiration in the neonatal intensive care unit
Author: Jorge, Joao
ISNI:       0000 0004 7653 3882
Awarding Body: University of Oxford
Current Institution: University of Oxford
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Camera-based monitoring is an emerging sensing modality which has attracted widespread interest due to its potential to offer comfortable and concurrent measurements of the cardiorespiratory vital signs: the cardiac pulse rate, respiratory rate and peripheral capillary oxygen saturation. In recent years, several methods have been proposed for the processing of visible and infrared video data and subsequent extraction of the respiratory rhythm. However, due to the low signal-to-noise ratio (SNR) of these signals, their application has so far been limited to adult subjects and controlled research environments. In this thesis, we aim to improve the signal processing aspects of non-contact respiratory measurements by developing a novel technique for extracting this signal in a clinical setting. The main contribution of this work is a multi-channel source separation method which combines information from multiple image domains and uses a priori information about the distinctive temporal structure of the signal of interest to improve the quality of the respiratory signal extracted. Considering the neonatal breathing component as an extreme case of poor SNR in video signals, we present a case study of the extraction of respiratory signals from video signals recorded from a CCD camera placed above an incubator inside which critically-ill infants are nursed. Evaluation results were obtained from recordings on 30 neonatal patients nursed in the Neonatal Intensive Care Unit at the John Radcliffe Hospital, Oxford, UK. The respiratory signals extracted using the methodology described compare favourably against those derived from direct and indirect sources during periods of manual breath counts. In addition, we show that a classification algorithm which uses features of these signals to identify periods of cessation of breathing can successfully distinguish between true and artefactual decreases in respiratory rate in a range of critically low values of this vital sign.
Supervisor: Tarassenko, Lionel Sponsor: Fundacao para a Ciencia e Tecnologia ; Portugal ; Engineering and Physical Sciences Research Council ; RCUK Digital Economy Programme
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Biomedical engineering