Meeting Banner
Abstract #4102

Segmented Total Disease Volume and Associated Apparent Diffusion Coefficient on Whole Body Diffusion-Weighted MRI Show Good Interobserver Agreement

Matthew D. Blackledge1, Nina Tunariu1, David J. Collins1, Martin O. Leach1, Dow-Mu Koh1

1CR-UK and EPSRC Cancer Imaging Centre, Institute of Cancer Research and Royal Marsden Hospital, Sutton, United Kingdom

Whole body DWI (WBDWI) allows the duo quantification of the total diseased diffusion volume and associated global mean/ median apparent diffusion coefficient (ADC), which are promising response biomarkers. However, the interobserver agreement of this approach has not been previously established. We found excellent interobserver agreement for segmenting total diseased diffusion volume on WBDWI and very good interobserver agreement for global median ADC values. These quantitative indices are robust and could be investigated more widely in future clinical trials, especially for the assessment of metastatic bone disease.

Keywords

ability account accuracy addition agreement annual apparent application applied apply assessment association axial background bandwidth better biomedical body bone breast breathing burden cancer cancers clear clinical coefficient computed conditional confidence coronal critical currently density derive described detecting determine determined developing diagnostic difficult diffusion disease diseased diseases displayed england estimating estimation evaluation examination experience feasible field free frequency functions funding generally generated global good grant health highly hospital initial institute intraclass kingdom leach likely logarithmic logarithms maps marrow martin matrix maximize maximized median medical meeting metastatic methodology metrics model modes natural observer ovary patient patients pixel plots presence prostate prove quantitative radiological radiologists radiology random reader readers receiver recorded relatively reported respectively robust scale scientist sector segmentation segmented selected self sensitivity skull slice soft software staging stations statistical stir studies suggesting suppressed suppression system tendency threshold throughout tissue treatment underwent unreliable variability variable variation vertex view volume volumes whole wider workstation years