Whole-body magnetic resonance imaging at 0.05 Tesla

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Artificial Intelligence in Healthcare

Background:

• Global magnetic resonance imaging (MRI) accessibility is limited and uneven, despite decades of advancement.
• Early MRI focused on low fields, but progress shifted to high-field superconducting magnets after 1983.
• Existing MRI technology faces challenges with cost, infrastructure, and accessibility in diverse healthcare settings.

Purpose of the Study:

• To develop an affordable, accessible whole-body MRI scanner.
• To overcome limitations of traditional high-field MRI systems.
• To enhance image quality in ultra-low-field MRI using deep learning.

Main Methods:

• Developed a whole-body scanner using a permanent 0.05 Tesla magnet.
• Utilized deep learning for electromagnetic interference elimination.
• Implemented 3D deep learning reconstruction leveraging high-field MRI data for image enhancement.

Main Results:

• Demonstrated a functional whole-body MRI scanner operating on standard power without special shielding.
• Showcased the scanner's applicability for imaging diverse anatomical structures.
• Achieved improved image quality through deep learning reconstruction techniques.

Conclusions:

• The developed ultra-low-field MRI scanner offers a potential solution for increasing global MRI accessibility.
• Deep learning integration significantly enhances image quality, making low-field MRI clinically viable.
• This technology paves the way for affordable, powerful MRI systems in underserved healthcare environments worldwide.