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

Area of Science:

• Medical Imaging
• Artificial Intelligence in Medicine
• Cardiovascular Imaging

Background:

• Carotid atherosclerotic plaques are a major cause of stroke.
• Accurate imaging of plaque characteristics is crucial for risk stratification.
• Conventional MRI techniques can be time-consuming and may have limitations in image quality.

Purpose of the Study:

• To evaluate the image quality and diagnostic performance of fast magnetic resonance imaging with deep learning-based reconstruction (MRI-DLR) for carotid atherosclerotic plaques.
• To compare MRI-DLR with conventional MRI (MRI-C) and fast MRI without deep learning reconstruction (MRI-Fast).

Main Methods:

• Sixty-nine patients underwent carotid plaque imaging using a 3.0T MRI system.
• Image quality was assessed objectively (SNR, CNR) and subjectively.
• MRI-DLR was compared against MRI-C and MRI-Fast.

Main Results:

• MRI-DLR and MRI-Fast reduced scanning time by approximately 63.2% compared to MRI-C.
• MRI-DLR demonstrated significantly higher SNR and CNR than MRI-Fast and comparable or superior SNR/CNR to MRI-C.
• Subjective image quality and diagnostic performance for plaque components (hemorrhage, lipid nuclei, etc.) were significantly higher with MRI-DLR.

Conclusions:

• MRI-DLR offers superior image quality and diagnostic performance for carotid atherosclerotic plaques.
• The technique significantly reduces scanning time, improving diagnostic accuracy and patient experience.
• This advancement aids in the diagnosis and treatment of stroke by providing better plaque characterization.