Deep learning-based reconstruction improves MRI image quality and diagnostic performance for carotid atherosclerotic plaques

Objective To evaluate the image quality and diagnostic performance of fast magnetic resonance imaging with deep learning-based reconstruction (MRI DLR ) for carotid atherosclerotic plaques. We compared MRI DLR to conventional MRI (MRI C ) and fast MRI without deep learning-based reconstruction (MRI Fast ). Methods Sixty-nine patients were recruited between May 2022 and April 2024. Imaging was performed using a 3.0 T MRI system (SIGNA™ Architect, GE Healthcare) with a 19-channel head–neck coil. Image quality was assessed objectively [measuring signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR)], subjectively (evaluating overall image quality, noise, contrast, artifacts, sharpness and diagnostic performance). All statistical analyses were performed using R software version 4.4.2, with agreement among results assessed using Cohen’s κ . Results The average scanning time was reduced by 12.4 min (approximately 63.2%) for MRI DLR and MRI Fast compared to MRI C . The SNR of T 1 and T 2 sequences for MRI DLR and MRI C showed significantly higher than MRI Fast ( p < 0.05), and the SNR of T 1 and T 2 sequences for MRI DLR was significantly higher than MRI C . The SNR of PD sequences for MRI DLR and MRI C showed significantly higher than MRI Fast ( p < 0.05), and there was not significantly different between MRI DLR and MRI C ( p > 0.05). The CNR of T 1 , T 2, and PD sequences for MRI DLR and MRI C showed significantly higher than MRI Fast ( p < 0.05), and there was not significantly different between MRI DLR and MRI C ( p > 0.05). Subjective image quality was significantly higher for MRI DLR than MRI C and MRI Fast ( p < 0.05), and MRI C was higher than MRI Fast ( p < 0.05). Comparing with MRI C and MRI Fast , MRI DLR was more effective in diagnosing hemorrhage, lipid nuclei, calcification, fiber cap disruption, and vulnerable plaques, and showed greater concordance with surgical pathology findings. Conclusion MRI DLR can offer superior image quality and diagnostic performance for carotid atherosclerotic plaques while reducing scanning time. So it enhances diagnostic accuracy and lessens the examination burden on patients which will contribute to the diagnosis and treatment of stroke.