Meeting Banner
Abstract #4106

Post-Hoc Spatial Registration of Multiple B-Value Diffusion Weighted Images Improves Test-Retest Reliability of ADC Measurements in Solid Tumors in the Body

Alexandre Coimbra1, Jill Fredrickson1, Roel Funke1, Daniel Chen1, Alex de Crespigny1

1Oncology Early Clinical Development, Genentech, Inc., South San Francisco, CA, United States

Accurate assessment of tissue apparent diffusion coefficient (ADC) requires acquisition and analysis of multiple b-value diffusion weighted images (DWI). One important source of variability in this ADC assessment is spatial misregistration across the multiple DWIs, particularly those at high b-values. The current work describes a simple post-hoc automated approach for spatial registration of multiple b-value DWIs. Applied to test-retest data set from a cohort of cancer patients, this approach enhanced repeatability of ADC measurements.

Abstract PDF E-Poster Video

Keywords

albeit analyses applied approved array assessment automated axes bland blue board body cancer clinical coefficient cohort coils correlation dashed days defined depicts describes develop development diffusion done drug early eddy either example free fusion future ground house improves institution interval intervention iteratively larger lowing manually maps matrix measured median meter metric minimize model nets original outlines overall overlaid panels part particularly patient patients plots post procedure processing proposed pulse receiving reed registered registration reliability repeat repeatability reports requires retest review scanned sensitivity simple slice slices smaller solid spatial statistically summary systematically therapeutic tissue treatment trial tumor twice variability variation week