Our Scoliosis research is led by Prof Emma Clark. Scoliosis (spinal curvature) has rarely been looked at in entire populations of people. To address this we have developed and validated a manual method to measure scoliosis from total-body bone density (DXA) scans[1], funded by the British Scoliosis Research Foundation.  

We have used this new method in the Avon Longitudinal Study of Parents and Children (ALSPAC) to identify predictors of onset of Adolescent Idiopathic Scoliosis (AIS). We have found that those who have less physical ability at aged 18 months, who do less physical activity at 9 and 11 years old[2], or who have reduced fat mass and muscle mass[3] are more likely to develop scoliosis between age 9 and 15. We have also shown that even small spinal curves at 15 years old are associated with pain and time off school at 18 years old [4].

We are currently working with colleagues from the University of Oxford ( to use machine learning techniques to automate the measurement of scoliosis from total body DXA scans[5]. We will be using thus automation to investigate the prevalence and epidemiology of adult degenerative scoliosis using UK Biobank as part of the Augment study funded by the Welcome Trust.

[1] Taylor HJ, Harding I, Hutchinson J, Nelson I, Blom A, Tobias JH, Clark EM (2013) Identifying scoliosis in population-based cohorts: Development and validation of a novel method based on total body DXA scans. Calcified Tissue International 92(6):539-547

[2] Tobias JH, Fairbank J, Harding I,Taylor HJ, Clark EM (2018). Association between physical activity and scoliosis: A prospective cohort study. International Journal of Epidemiology doi: 10.1093/ije/dyy268

[3] Clark EM, Taylor JH, Harding I et al (5 additional authors) (2014) Association between components of body composition and scoliosis: A prospective cohort study reporting differences identifiable before the onset of scoliosis. J Bone Miner Res 29(8):1729-1736.

[4] Clark EM, Tobias JH, Fairbank J (2016) The impact of small spinal curves in adolescents that have not presented to secondary care: a population-based cohort study. Spine 41(10):E611-E617.

[5] Jamaludin A, Kadir T, Clark EM, Zisserman A (2018) Predicting scoliosis in DXA scans using intermediate representations. 5th International Workshop and Challenge on Computational Methods and Clinical Applications for Spine Imaging. Post-proceedings, Springer’s Lecture Notes in Computer Science, vol 11397, pg 15-28, DOI 10.1007/978-3-030-13736-6

Edit this page