Abstract #3434

Comparison of six different diffusion weighted MRI models in pancreatic cancer patients

Oliver J. Gurney-Champion^1,2, Remy Klaassen³, Martijn Froeling^1,4, Jaap Stoker¹, Johanna W. Wilmink^3,5, Arjan Bel², Hanneke W.M. van Laarhoven³, and Aart J. Nederveen¹

¹Radiology, Academic Medical Center, Amsterdam, Netherlands, ²Radiation Oncology, Academic Medical Center, Amsterdam, Netherlands, ³Medical Oncology, Academic Medical Center, Amsterdam, Netherlands, ⁴Radiology, University Medical Center Utrecht, Utrecht, Netherlands, ⁵Internal Medicine, Academic Medical Center, Amsterdam, Netherlands

We compare the performance of six models for diffusion weighted MRI (DWI) data signal intensity as a function of b-values. We obtained three DWI-datasets spread over two sessions from 9 pancreatic cancer patients. To characterize the fit parameters’ sensitivity to abnormalities, we compared the difference in fit parameter values between tumorous and healthy tissue. To characterize repeatability, we obtained the coefficient of variation. Every model had exactly one parameter that was significantly sensitive to tumorous tissue (p<0.05). From these parameters, the most sensitive parameter was f₂ from the tri-exponential model, whereas the best repeatable measure was the ADC from the mono-exponential model.

This abstract and the presentation materials are available to members only; a login is required.

Join Here