Encrypted login | home

Program Information

Distortion-Free 3D Diffusion MRI On An MRI-Guided Radiotherapy System for Longitudinal Tumor Response Assessment

no image available
Y Gao

Y Gao*, Y Yang , N Rangwala , M Cao , D Low , P Hu , UCLA School of Medicine, Los Angeles, CA


TU-AB-BRA-7 (Tuesday, August 2, 2016) 7:30 AM - 9:30 AM Room: Ballroom A

To develop a reliable, 3D distortion-free diffusion MRI technique for longitudinal tumor response assessment and MRI-guided adaptive radiotherapy(RT).

A diffusion prepared 3D turbo spin echo readout (DP-TSE) sequence was developed and compared with the conventional diffusion-weighted single-shot echo-planar-imaging (DW-ssEPI) sequence in a commercially available diffusion phantom, and one head-and-neck and one brain cancer patient on an MRI-guided RT system (ViewRay). In phantom study, the geometric fidelity was quantified as the ratio between the left-right (RL) and anterior-posterior (AP) dimension. Ten slices were measured on DP-TSE, DW-ssEPI and standard TSE images where the later was used as the geometric reference. ADC accuracy was verified at both 0°C (reference ADC available) and room temperature with a range of diffusivity between 0.35 and 2.0*10⁻³mm²/s. The ADC reproducibility was assessed based on 8 room-temperature measurements on 6 different days. In the pilot single-slice in-vivo study, CT images were used as the geometric reference, and ADC maps from both diffusion sequences were compared.

Distortion and susceptive-related artifact were severe in DW-ssEPI, with significantly lower RL/AP ratio (0.9579±0.0163) than DP-TSE (0.9990±0.0031) and TSE (0.9995±0.0031). ADCs from the two diffusion sequences both matched well with the vendor-provided values at 0°C; however DW-ssEPI fails to provide accurate ADC for high diffusivity vials at room temperature due to high noise level (10 times higher than DP-TSE). The DP-TSE sequence had excellent ADC reproducibility with <4% ADC variation among 8 separate measurements. In patient study, DP-TSE exhibited substantially improved geometric reliability. ROI analysis in ADC maps generated from DP-TSE and DW-ssEPI showed <5% difference where high b-value images were excluded from the latter approach due to excessive noise level.

A diffusion MRI sequence with excellent geometric fidelity, accurate and highly reproducible ADC measurements was proposed for longitudinal tumor response assessment using an MRI-guided RT system.

Funding Support, Disclosures, and Conflict of Interest: Yu Gao acknowledges research support from ViewRay.

Contact Email: