Impact of geometric correction on echo-planar imaging-based apparent diffusion coefficient maps for abdominal...

Area of Science:

• Radiotherapy Physics
• Medical Imaging
• Biomarker Development

Background:

• Apparent diffusion coefficient (ADC) from diffusion-weighted magnetic resonance imaging (DWI) is a potential radiotherapy biomarker.
• Echo-planar imaging (EPI) read-out in DWI offers speed but suffers from geometric inaccuracies.
• Geometric distortions can affect ADC values and tumor shape assessment in abdominal EPI-DWI.

Purpose of the Study:

• To investigate the influence of geometric distortions on ADCs within the gross tumor volume (GTV).
• To assess the impact of distortions on the GTV shape in abdominal EPI-DWI.
• To evaluate the necessity of distortion correction for ADC-based radiotherapy outcome prediction and dose painting.

Main Methods:

• Collected EPI-DWI scans from 21 patients on MRI-Sim and MRI-Linac systems.
• Applied off-resonance and eddy current corrections to ADC maps and deformed clinical GTVs.
• Quantified geometric changes using Mean Surface Distance (MSD), Hausdorff Distance (HD), and Dice Similarity Coefficient (DSC); calculated ADC values using a mono-exponential model.

Main Results:

• Median ADC change within the GTV after full distortion correction was minimal (1.3% for MRI-Sim, 1.5% for MRI-Linac).
• Additional eddy current correction had a small effect on both systems.
• Significant local-regional geometric changes in GTV were observed after off-resonance distortion correction (median MSD: 0.43 mm, median HD: 4.00 mm, median DSC: 0.93).

Conclusions:

• Distortion correction has a minor impact on ADC values, suggesting it may not be crucial for ADC-based outcome prediction in radiotherapy.
• However, substantial geometric alterations in GTV shape necessitate distortion correction for strategies relying on spatial ADC map information, such as dose painting.
• Phantom measurements validated the applied correction methods.