Meeting Banner
Abstract #0505

Characterizing Response in Head and Neck Cancer with Cluster Analysis of Multi-Parametric MRI Data

Marco Borri1, Maria A. Schmidt1, Ceri Powell2, Dow-Mu Koh, 13, Angela Riddell3, Kate Newbold2, Martin O. Leach1

1CR-UK and EPSRC Cancer Imaging Centre, Institute of Cancer Research and Royal Marsden, Sutton, Surrey, United Kingdom; 2Head and Neck Dept., The Royal Marsden NHS Foundation Trust, Sutton, Surrey, United Kingdom; 3Radiology Dept., The Royal Marsden NHS Foundation Trust, Sutton, Surrey, United Kingdom

One of the main challenges in the management of head and neck squamous cell carcinomas (HNSCC) is the inability to predict outcome because of tumour heterogeneity. We propose a method, based on cluster analysis, to robustly partition head and neck lesions in sub-regions using complementary functional information from DCE and DWI. This approach was able to spatially describe functional heterogeneity in a cohort of patients with histologically proven HNSCC undergoing radical chemoradiotherapy, and allowed identification of sub-regions of the tumours which were affected differently by treatment and their characterization with functional parameters.

Abstract PDF Presentation Video

Keywords

able advanced affected apparent applied approved around association axis become benefit best biological biomedical cancer central characteristics chemotherapy choice cited classification clearly clinical cluster clustering clusters committee complete component conformal consent contoured contrast control corrected cumulative curve cycles dept derived describe describing determines differently diffusion dimensional disease distortion distribution doses efficiency england enhanced ethics evaluate evolution expressed formed foundation functional functionality funding gadolinium global head health heterogeneity highlights house identification identified identify indicates induction initial intensify intensifying jump justifying kingdom largest leach lesion lesions likely local locally management maps model mutual neck necrotic nine nodes nonlinear occurring optimal outcome outcomes parametric partition partitioned patients percentage performance perfusion poor post predict preliminary principal processes propose protocol radiation radical radiology radiotherapy reducing reduction reflected registering representative resistant restricted risk royal section segmentation separated separating sharp simplest slice slices software solutions specified stratifying sugar surrey table target theory treatment trust type units validation volume