Reducing energy consumption in musculoskeletal MRI using shorter scan protocols, optimized magnet cooling patterns,...

Area of Science:

• Radiology
• Medical Imaging
• Energy Efficiency

Background:

• Rising energy costs in Europe impact radiology departments.
• MRI scanners are significant energy consumers.
• Optimizing MRI energy usage is crucial without compromising image quality or efficiency.

Purpose of the Study:

• Investigate MRI energy consumption.
• Identify strategies for improving energy efficiency in musculoskeletal MRI.
• Assess potential savings from protocol optimization, deep learning (DL) acceleration, and cooling system optimization.

Main Methods:

• Energy consumption measured on two 1.5-T MRI scanners (Aera, Sola) in Munich.
• Implemented and compared three levels of energy reduction measures against baseline.
• Used Wilcoxon signed-rank test with Bonferroni correction to analyze scan times and energy consumption.

Main Results:

• Deep learning (DL) reduced energy consumption by 72% and scan time by 71%.
• Economic protocols reduced energy consumption by 31% and scan time by 18%.
• Optimizing the cooling system (Eco power mode) reduced energy consumption by 30% during non-scanning periods, saving over €4881 annually per scanner.

Conclusions:

• Energy-saving strategies like economic protocols, DL-accelerated sequences, and optimized magnet cooling significantly reduce MRI scanner energy consumption.
• Adopting these strategies can lead to substantial cost reductions and environmental benefits for radiology practices.
• These optimizations are vital for sustainable and cost-effective medical imaging operations.