Stephanie B. Donaldson1,2, Catharine M.L. West3, Susan E. Davidson4, Bernadette M. Carrington5, Gillian Hutchison5, Steven P. Sourbron6, David L. Buckley2
1North Western Medical Physics, Christie Hospital NHS Foundation Trust, Manchester, UK; 2Imaging Science and Biomedical Engineering, University of Manchester, Manchester, UK; 3Academic Department of Radiation Oncology, University of Manchester, Manchester, UK; 4Department of Clinical Oncology, Christie Hospital NHS Foundation Trust, Manchester, UK; 5Department of Radiology, Christie Hospital NHS Foundation Trust, Manchester, UK; 6Josef Lissner Laboratory for Biomedical Imaging, Ludwig-Maximilian-University, Munich, Germany
The modified-Tofts tracer kinetic model is often used to analyse DCE-MRI data but its validity is dependent upon assumptions such as negligible mean transit time (MTT). We compared the modified-Tofts model with its parent model, a general 2-compartment exchange model (2CXM), using DCE-MRI data from 27 patients with cervical cancer. Fits obtained with the modified-Tofts model were poorer than those obtained with the 2CXM and comparable parameters were significantly different. MTT was non-negligible suggesting that the 2CXM is more suitable for the analysis of this DCE-MRI dataset.