A Deep Learning-Based De-Artifact Diffusion Model for Removing Motion Artifacts in Knee MRI.

Abstract

Background: Motion artifacts are common for knee MRI, which usually lead to rescanning. Effective removal of motion artifacts would be clinically useful.

Purpose: To construct an effective deep learning-based model to remove motion artifacts for knee MRI using real-world data.

Study type: Retrospective.

Subjects: Model construction: 90 consecutive patients (1997 2D slices) who had knee MRI images with motion artifacts paired with immediately rescanned images without artifacts served as ground truth. Internal test dataset: 25 patients (795 slices) from another period; external test dataset: 39 patients (813 slices) from another hospital.

Field strength/sequence: 3-T/1.5-T knee MRI with T1-weighted imaging, T2-weighted imaging, and proton-weighted imaging.

Assessment: A deep learning-based supervised conditional diffusion model was constructed. Objective metrics (root mean square error [RMSE], peak signal-to-noise ratio [PSNR], structural similarity [SSIM]) and subjective ratings were used for image quality assessment, which were compared with three other algorithms (enhanced super-resolution [ESR], enhanced deep super-resolution, and ESR using a generative adversarial network). Diagnostic performance of the output images was compared with the rescanned images.

Statistical tests: The Kappa Test, Pearson chi-square test, Fredman's rank-sum test, and the marginal homogeneity test. A p value < 0.05 was considered statistically significant.

Results: Subjective ratings showed significant improvements in the output images compared to the input, with no significant difference from the ground truth. The constructed method demonstrated the smallest RMSE (11.44  $\pm$  5.47 in the validation cohort; 13.95  $\pm$  4.32 in the external test cohort), the largest PSNR (27.61  $\pm$  3.20 in the validation cohort; 25.64  $\pm$  2.67 in the external test cohort) and SSIM (0.97  $\pm$  0.04 in the validation cohort; 0.94  $\pm$  0.04 in the external test cohort) compared to the other three algorithms. The output images achieved comparable diagnostic capability as the ground truth for multiple anatomical structures.

Data conclusion: The constructed model exhibited feasibility and effectiveness, and outperformed multiple other algorithms for removing motion artifacts in knee MRI.

Evidence level: Level 3.

Technical efficacy: Stage 2.