Meeting Banner
Abstract #1548

Improving the Accuracy of DCE-MRI-Based Prediction of Bevacizumab- And FOLFOX6-Induced CRC Liver Metastasis Shrinkage

Saada A M Abujarada1, 2, James P B O'Connor1, 3, Chris J. Rose1, 2

1University of Manchester Biomedical Imaging Institute, Manchester, Greater Manchester, United Kingdom; 2University of Manchester Academic Health Science Centre, Manchester, Greater Manchester, United Kingdom; 3Department of Radiology, Christie Hospital NHS Trust, Manchester, Greater Manchester, United Kingdom

MRI-derived biomarkers, measured before treatment, may predict tumor response. Such research tends to use linear regression that may overlook differences in biomarker reliability (interscan repeatability). Where reliability is considered, it is often quantified separately to the regression task, rather than within the same conceptual framework. We describe a novel nonlinear model that explicitly models biomarker reliability and compare its predictions to those from a linear errors-in-variables model in a study of liver metastases treated with bevacizumab and FOLFOX6. While the median prediction error of the methods is equal, our method makes substantially fewer large errors.

Abstract PDF

Keywords

ability academic according account accounts accuracy accurate actual agent aggregation allow apply assessing attend believe best biomedical bland bootstrap cancer chosen city clear combined come committee common conceptual confirm consent control corp costs cycle developed deviation disrupting distinct distributions drawn emerging enhancing ensemble error errors ethics evaluate example excel explained explicitly fewer find fitted fraction framework function functions gnostic good greater health heterogeneity hows independent inspect intro kingdom learning least leave linear liver location logarithms made magnitude makes maps measured median metastasis minim minimal missing model modeling models nonlinear optimal particular patient plot poor predicted prediction predictions predictive predictor prior problem procedures quality quantified reason received refer refs regression reliability reliable remaining repeatability response rose sample samples shrinkage software spatial springer stratify stump stumps subset system take takes tasks term terminal terms trained transfer transform transl treatment tree tumor unction understood underwent unreliable variable variables variance vary volume weeks wise wolfram