Abstract #1853
Radiomics analysis of multi-parametric MRI in human brain tumours
Martin Lowry 1 , Lawrence Kenning 2 , and Lindsay W Turnbull 1
1
Centre for MR Investigations, Hull York
Medical School at University of Hull, Hull, East
Yorkshire, United Kingdom,
2
Centre
for MR Investigations, University of Hull, Hull, East
Yorkshire, United Kingdom
Multi-parametric MRI produces large quantities of data
in which parameter associations to tissue
characteristics may be hidden. We applied the data
mining approach of Radiomics to investigate the
information content of a pre-operative multi-parametric
MR study of gliomas. The results demonstrate redundancy
of some parameters but also that DCE, DSC, and DTI can
provide complimentary parameters. Several distinct
parameter profiles were associated with high grade
lesions but only one with low grade lesions despite
variation in known prognostic genetic markers amongst
the tumours. The approach shows potential for
identifying important associations that could influence
treatment or predict survival.
This abstract and the presentation materials are available to members only;
a login is required.
Join Here