Hypothesis: Approximately 12,000 tracheostomies are performed annually in the UK. However, complications are commonplace, with tracheostomy tube sizing being an important factor in mortality and morbidity as highlighted in the 2014 National Confidential Enquiry into Patient Outcome and Death (NCEPOD) report "Tracheostomy care: On the right trach?". There is a lack of normative data on the variation of pretracheal soft tissue depth, tracheal dimensions and tracheal/spinal morphology with age and sex for the adult population in the United Kingdom. We hypothesise that the degree of variation in neck dimensions and morphology for the adult population in the UK is greater than the variety of commercially available tracheostomy tube dimensions.
Aims: The study will utilise machine learning and computer vision to analyse normal tracheal course, morphology and its relationship with the skin contour and spinal anatomy using 1000 neck and thorax CT scans of adult patients conducted across NHS trusts in the UK. The trachea tends to follow the course of the spine rather than the direction of skin contour. Therefore, when spinal curvature is more pronounced, the trachea may lie deeper in the root of the neck than anticipated. Tateyama et al have suggested that advanced cervical lordosis angle is involved in misalignment of tracheostomies. We aim to establish predictable patterns of neck morphology that will allow selection of the correct tracheostomy tube sizing based upon easily identifiable anatomical landmarks and measurements e.g. neck length, skin depth, cervical lordosis etc. We aim to reduce the NHS cost burden by following the NHS principal of GIRFT 'getting it right the first time' and fitting the correct tube on the first attempt with the algorithm that we will produce from this work, and remove the concept of trial-and-error in fitting tracheostomy tubes.
Objectives and Methods: We will conduct a national multicentre retrospective study. Ten collaborators, each from a different NHS hospital have collected data for 100 patient CT neck and thorax scans each. Tracheal course, morphology and its relationship with spinal anatomy will be explored via visual representation by superimposing lines outlining the skin contour, the anterior and posterior tracheal walls as well as the anterior border of the cervical/thoracic spine, from approximately the level of hyoid to level of the clavicles. The hyoid bone, manubrium and C7 will also be identified and presented. Identification of these features will be semi-automated, adapting existing computer vision software developed to identify osteoporotic vertebral fractures in CT of the thoracic and lumbar spine (www.stopfrac.org). This data set will then be analysed using statistical shape and appearance modelling techniques to generate models of tracheal anatomy that summarise normative shape variation and predict optimal tracheostomy tube dimensions.
Potential Outcomes: 1) Normative data on tracheal morphology in the study population, allowing identification of any shortfall in the dimensions of commercially available trachesotomy tubes. 2) A computer vision model that can analyse neck CT images to predict tracheal morphology and optimal tracheostomy tube dimensions. If successful, we will seek engagement with medical device companies to address any shortfall in available tracheostomy tube dimensions, and to commercialise the predictive modelling software, potentially through existing collaborations with partners such as Optasia Medical Ltd. (www.optasiamedical.com).
Clinical impact: This project will advance the knowledge on neck anatomy in relation to the trachea. This will benefit the surgeons and anaesthetists who operate on the neck and in particular for those who perform tracheostomies. It will guide and help design better positioning of the tracheal incision.
This project will help guide the operating surgeon / anaesthetist in choosing the appropriate tube for the patient. There is huge wastage in the trial-and-error principal followed at present in the NHS due to a lack of knowledge on this subject; each tube costs on average £150 and multiple tubes are tried per patient (annual cost to NHS ~£20M). This project has the potential in making significant cost savings in reducing wastage, whilst also improving patient toleration and comfort of tracheostomy tubes in the peri- and post-operative period.
This project will advance our knowledge in understanding variability in anatomy and its consequences in health product design. This will allow us to work collaboratively in improving or even redesigning tracheostomy tubes to benefit patients by saving lives without increasing their risk for complications.
Training: The student will gain extensive experience of state-of-the-art machine vision algorithm development, of medical image analysis, and of working in a collaborative environment involving academic and health-care partners. Dr Khwaja is a Consultant Laryngologist and expert on ENT, and will lead on the clinical aspects of the project. Professor Brass is Professor of Bioinformatics and will lead on data analysis. Dr Bromiley is an expert in computer vision and medical image analysis, with experience on using statistical shape modelling to identify musculoskeletal structures in general and spinal structures in particular, and will lead on the image analysis aspects.
Entry Requirements
Applicants must have obtained, or be about to obtain, at least an upper second class honours degree (or equivalent) in a relevant subject.
How to Apply
To be considered for this project you MUST submit a formal online application form. Please select EPSRC PhD Programme on the online application form. For information on how to apply for this project, please visit the Faculty of Biology, Medicine and Health Doctoral Academy website (https://www.bmh.manchester.ac.uk/study/research/apply/).
Applications must be submitted by the deadline, as late applications will not be considered. Incomplete applications will not be considered. Please ensure your application is complete and includes all required documentation before submission.
Applicants interested in this project should make direct contact with the Primary Supervisor to arrange to discuss the project further as soon as possible.
Equality, Diversity and Inclusion
Equality, diversity and inclusion is fundamental to the success of The University of Manchester, and is at the heart of all of our activities. The full Equality, diversity and inclusion statement can be found on the website https://www.bmh.manchester.ac.uk/study/research/apply/equality-diversity-inclusion/