LAUSR

Purpose The aim of this study is to propose a protocol for DWI QA being easy to implement while still providing quantitative information on relevant parameters for the scanner QC as well as for clinical applications. The measured quantity is the apparent diffusion coefficient (ADC) and its dependence from gradient diffusion direction (Gdd) and phase encoding direction (PED). Methods A spherical water phantom (14 cm diameter), designed to allow fast and easy measurement of solution temperature, is proposed. The phantom is filled with doped aqueous solution (ADC value certified) and placed at isocentre of scanner. The protocol consists of five set of acquisition of 19 slices, repeated for three orthogonal Gdd. First and second acquisitions are identical, the others differ for the PED and acquisition plane. Each slice (for each Gdd) is acquired with six different b-values varying from 0 s/mm2 to 1000 s/mm2. From each acquisition and Gdd, ADC map of the phantom is generated with monoexponential fit, corrected for temperature and ADC profile along three orthogonal direction is extracted. From first two acquisitions short term reproducibility and SNR are evaluated; second and third acquisitions are used to detect phase encoding direction effects, lasts three maps are used to detect acquisition plane effects on ADC (details in Table 1 ). Results A preliminary test of the protocol on 1.5T MRI scanner demonstrated its applicability to routine QA. Relative difference between on-centre and periphery ADC values varies from −5% and +3% and phase encoding effect is less than 0.5%. Gdd variations are around +1%. Acquisition plane differences are less than 1% at isocentre and of the order of 5% in periphery. SNR at isocenter is 550 and short term variation is 0.2%. Conclusions The proposed protocol provides a simple QA procedure for DWI application. Selected indexes provide quantitative information on many relevant acquisition parameters.