Physically implausible signals as a quantitative quality assessment metric in prostate diffusion-weighted MR imaging

Area of Science:

• Radiology
• Medical Imaging
• Biophysics

Background:

• Diffusion-weighted imaging (DWI) is crucial for prostate cancer detection.
• Artifacts in DWI can compromise image interpretation and diagnostic accuracy.
• Quantifying artifacts like physically implausible diffusion signal (PIDS) is essential for quality assessment.

Purpose of the Study:

• To quantitatively assess artifacts in prostate DWI using PIDS.
• To compare PIDS levels in prostate MRI scans acquired with and without an endorectal coil (ERC).
• To evaluate the distribution of PIDS across different prostate zones.

Main Methods:

• Retrospective analysis of 80 prostate MRI scans (40 with ERC, 40 without).
• PIDS defined as voxels with signal increase vs. b-value or ADC > 3.0 μm²/ms.
• In-house MATLAB program used for PIDS calculation; statistical analysis (t-test, ANOVA) performed.

Main Results:

• Average PIDS was 17.1% with ERC and 22.2% without ERC, with no significant difference (p=0.14).
• PIDS levels did not significantly differ between ERC and NERC cohorts overall.
• The mid-prostate zone in ERC scans showed significantly lower PIDS (13.8%) compared to the apex and base.

Conclusions:

• PIDS mapping offers a quantitative, spatially resolved quality assessment for prostate DWI.
• Overall PIDS levels are comparable with or without an ERC, but focal PIDS can be high in specific zones.
• PIDS assessment can inform radiologists and guide the development of improved DWI sequences.