Computer-aided Detection of Osteoporotic Vertebral Fractures in Clinical Images Using Convolutional Neural Network Constrained Local Models (CNN-CLMs)

   Faculty of Biology, Medicine and Health

  , ,  Applications accepted all year round  Self-Funded PhD Students Only

About the Project

Osteoporosis is a common, degenerative skeletal disorder that increases the risk of fractures and causes significant morbidity and mortality. Vertebral fractures (VFs) are often an early manifestation of the disease but are significantly under-diagnosed in clinical practice. The University of Manchester (UoM) is currently collaborating with Optasia Medical Ltd. and the Manchester University NHS Foundation Trust to develop ASPIRE(TM), an out-sourced radiology reporting service for VFs. This uses UoM appearance-modelling technology to assist radiologists by identifying vertebrae in clinical images, measuring their shape, and classifying fractures. However, in recent years, deep learning systems based on convolutional neural networks (CNNs) have shown superior performance on many image analysis tasks.

This project will develop and validate CNNs that can identify VFs in clinical images and could be incorporated into ASPIRE(TM) to improve accuracy and efficiency. Regional CNN regressors that can identify vertebral body outlines will be combined with statistical shape models of the spine, the latter providing a constraint that allows training on relatively small numbers (~1000s) of images whilst dealing with the highly repetitive nature of spinal structures. Through the course of the ASPIRE(TM) project we have acquired a database of over 2000 CT image volumes, each with manual annotations of the vertebral body outlines and diagnoses of spinal pathology. The algorithms will be trained, tested and validated using these images, allowing comparison to the accuracy of our current technology. The aim will be to improve the accuracy and efficiency of the technology used in ASPIRE(TM), reducing the cost of the service to the NHS.

The student will join a well-established, interdisciplinary team including academic, clinical and industrial partners. They will gain extensive knowledge of state-of-the-art computer vision algorithm development for clinical imaging problems. The student will have the opportunity to gain experience of working with an industrial partner, and the translational skills required to transfer research results between the academic and industrial environments. They will also have the opportunity to analyse results from the on-going feasibility studies of ASPIRE(TM), and to feed into an associated health economic study. Finally, they will work with our clinical partners from the Manchester Royal Infirmary and gain extensive knowledge of osteoporosis, particularly the clinical imaging, diagnosis and epidemiology of vertebral fractures and other spinal pathologies.

Entry Requirements

Candidates are expected to hold (or be about to obtain) a minimum upper second class honours degree (or equivalent) in a related area/subject. Candidates with previous laboratory experience are particularly encouraged to apply.

How To Apply

For information on how to apply for this project, please visit the Faculty of Biology, Medicine and Health Doctoral Academy website ( Informal enquiries may be made directly to the primary supervisor. On the online application form select the appropriate subject title.

For international students, we also offer a unique 4 year PhD programme that gives you the opportunity to undertake an accredited Teaching Certificate whilst carrying out an independent research project across a range of biological, medical and health sciences.

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”

Biological Sciences (4) Computer Science (8) Mathematics (25) Medicine (26)

Funding Notes

Applications are invited from self-funded students. This project fee band will be confirmed with the supervisor . Details of our different fee bands can be found on our website (View Website).


C. Lindner, P.A. Bromiley, M.C. Ionita and T.F. Cootes. Robust and Accurate Shape Model Matching using Random Forest Regression Voting. IEEE Transactions on Pattern Analysis and Machine Intelligence 37(9) p. 1862-1874, 2015.

P.A. Bromiley, J. Adams and T.F. Cootes. Localisation of Vertebrae on DXA Images using Constrained Local Models with Random Forest Regression Voting. Lecture Notes in Computational Vision and Biomechanics 20, p. 159-171, 2015.

P.A. Bromiley, E.P. Kariki, J.E. Adams and T.F. Cootes. Fully Automatic Localisation of Vertebrae in CT images using Random Forest Regression Voting. Lecture Notes in Computer Science 10182. p. 51-63, 2017.

P.A. Bromiley, E.P. Kariki, J.E. Adams and T.F. Cootes. Classification of Osteoporotic Vertebral Fractures using Shape and Appearance Modelling. Lecture Notes in Computer Science 10734, p. 133-147, 2018.

J. Staal, E. Kariki, R. Hyatt, M.K. Javaid, E. Russell, T. O'Neill, K. Poole, D. Chappell, R. Rajak. Multi-site opportunistic diagnosis of vertebral fragility fractures in computed tomography scans. Osteoporosis International 29(S2) p. S646, 2018.

Register your interest for this project