Meeting Banner
Abstract #4476

Combining Multi-Centre Conventional and Diffusion MR Texture for the Characterisation of Childhood Brain Tumours

S. Tantisatirapong 1 , N. P. Davies 1,2 , D. Rodriguez 3 , L. Abernethy 4 , D. P. Auer 3,5 , C. A. Clark 6,7 , R. Grundy 3,5 , T. Jaspan 5 , D. Hargrave 7 , L. MacPherson 2 , M. O. Leach 8 , G. S. Payne 8 , B. L. Pizer 4 , S. Bailey 9 , A. C. Peet 1,10 , and T. N. Arvanitis 10,11

1 University of Birmingham, Edgbaston, Birmingham, United Kingdom, 2 University Hospitals Birmingham NHS Foundation Trust, Birmingham, United Kingdom, 3 University of Nottingham, Nottingham, United Kingdom, 4 Alder Hey Childrens NHS Foundation Trust, Liverpool, United Kingdom, 5 University Hospital Nottingham, Nottingham, United Kingdom, 6 University College London, London, United Kingdom, 7 Great Ormond Street Hospital, London, United Kingdom, 8 The Institute of Cancer Research and Royal Marsden Hospital, Sutton, United Kingdom, 9 Northern Institute for Cancer Research, Newcastle University, Newcastle upon Tyne, United Kingdom, 10 Birmingham Childrens Hospital NHS Foundation Trust, Birmingham, United Kingdom, 11 Institute of Digital Healthcare, WMG, University of Warwick, Coventry, United Kingdom

This paper presents integration of multimodal MR image based texture analysis which take advantage of complementary information derived from structural and diffusion MR images. A supervised machine learning approach is used to achieve image analysis based on textural features from individual image types of T2, T1-post contrast and ADC, as well their combination, in order to characterize the multicenter dataset of the most common pediatric brain tumors; medulloblastomas (MB), pilocytic astrocytomas (PA), and ependymomas (EP).

This abstract and the presentation materials are available to members only; a login is required.

Join Here