Can we predict motion artifacts in clinical MRI before the scan completes?

Area of Science:

• Medical Imaging
• Radiology
• Magnetic Resonance Imaging

Background:

• Subject motion is a common issue in clinical Magnetic Resonance Imaging (MRI).
• Motion artifacts frequently lead to scan corruption and necessitate repeat acquisitions, causing inefficiency.
• Current methods for artifact detection often occur post-acquisition, delaying corrective actions.

Purpose of the Study:

• To develop a method for predicting MRI artifact scores from partial acquisitions.
• To enable early detection and potential abortion of corrupted scans during the acquisition process.
• To reduce scan time and improve patient throughput in clinical MRI settings.

Main Methods:

• Utilizing partial multi-shot, multi-slice MRI data.
• Developing a predictive model to estimate artifact scores.
• Implementing a real-time or near-real-time scoring system.

Main Results:

• Demonstrated the feasibility of predicting artifact scores using partial acquisition data.
• Showcased the potential for early identification of motion-corrupted scans.
• Quantified the expected improvement in scan efficiency.

Conclusions:

• Predicting artifact scores from partial MRI data is a viable strategy to mitigate motion-related inefficiencies.
• This approach can guide operators to abort scans prematurely, saving time and resources.
• Further validation in diverse clinical settings is warranted to optimize its application.