Abstract #4765

Retrospective ADC correction of gradient nonlinearity errors in multi-center breast DWI trials: ACRIN6698 multi-platform feasibility study

Dariya Malyarenko¹, David C Newitt², Lisa J Wilmes², Ek Tsoon Tan³, Luca Marinelli⁴, Ajit Devaraj⁵, Johannes M Peeters⁶, Shivraman Giri⁷, Axel vom Endt⁸, Nola Hylton², Savannah Partridge⁹, and Thomas L Chenevert¹

¹Radiology, University of Michigan, Ann Arbor, MI, United States, ²Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, United States, ³Hospital for Special Surgery, New York, NY, United States, ⁴GE Global Research, Niskayuna, NY, United States, ⁵Philips Research North America, Cambridge, MA, United States, ⁶Philips MR Clinical Science, Best, Netherlands, ⁷Siemens Medical Solutions, Boston, MA, United States, ⁸Siemens Healthcare GmbH, Erlangen, Germany, ⁹Radiology, University of Washington, Seattle, WA, United States

Multi-site, multi-platform clinical oncology trials seek to enhance quantitative utility of the apparent diffusion coefficient (ADC) metric by reducing technical cross-platform variability due to systematic gradient nonlinearity (GNL). Here we test feasibility of retrospective GNL correction implementation for a representative subset of subjects and systems from the ACRIN6698 breast cancer therapy response trial. GNL ADC correction based on previously developed formalism is demonstrated for trace-DWI DICOM using system-specific gradient-channel fields derived from vendor-provided spherical harmonic tables. Implemented correction substantially improves precision and removes ADC bias for DWI QC phantoms, and markedly changes ADC histogram percentiles for solid breast tumors.

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

Join Here