Since the inception of magnetic resonance imaging (MRI) in medical imaging, innovations in MRI have been driven by technological achievements from a wide array of scientific fields. This trend continues to this day, as current and future innovations seek to draw from different areas of advancement in hardware and software capabilities.1
Innovations in magnetics
Over decades of effort to improve efficiency and image quality in MRI technology, machines with the capacity to support increasingly strong static magnetic fields have been developed. More powerful machines have made the achievement of a higher signal-to-noise ratio (SNR) possible, a metric that measures efficiency of image acquisition, image resolution, and detection of pathology. But simply turning up the static magnetic field strength of an MRI machine comes with risks, and scientists continue to search for new technologies to optimize the balance between benefit and risk for patients and continue to advance MRI machine technology.2
Metamaterials in MRI
In recent years, experts at the Boston University Photonics Center created a device composed of a new magnetic metamaterial designed to interact with the magnetic field within the MRI machine to improve overall scan efficiency.2 This new artificially engineered metamaterial is an array of metallic helical unit cells with collective resonant modes that are excited by and enhance the radiofrequency magnetic field in MRI, leading to increased SNR.2
The device may one day allow using a lower strength magnetic field in MRI scanners. This development marks a possible future for device technology that could enhance the efficiency of MRI scanners while using a magnetic field significantly lower than the machines of today.2
Artificial intelligence (AI)
Accelerating the speed of image acquisition in MRI may finally be achievable through AI technology, utilizing deep learning image reconstruction. By accessing a network of open-source data MRI scans, AI technology has the capability to cut though extraneous raw data, creating images that match the clarity and diagnostic accuracy of current technology.3
A recent study done by the NYU Grossman School of Medicine and Facebook AI examined the use of deep learning image reconstruction to reduce MRI examination time. The findings of the study have suggested the use of deep learning in MRI may one day create high-quality images faster and with less data than traditional MRI.3
The study marks another step toward clinical acceptance of the use of AI in MRI, a technology that has potential to advance clinicians’ ability to reach diagnosis faster and with reliable accuracy.3
Other research into AI applications in MRI suggest that, while it may not be feasible to eliminate the use of contrast agents, it may be possible to achieve a dose reduction with the assistance of AI.4
References
- Viard A, Eustache F, Segobin S. History of magnetic resonance imaging: a trip down memory lane. Neuroscience. 2021;474:3-13.
- Duan G, Zhao X, Anderson SW, Zhang X. Boosting magnetic resonance imaging signal-to-noise ratio using magnetic metamaterials. Commun Phys. 2019;2(1):35. doi: 10.1038/s42005-019-0135-7.
- Recht M, Zbontar J, Sodickson DK, et al. Using deep learning to accelerate knee MRI at 3 T: results of an interchangeability study. AJR Am J Roentgenol. 2020;215:1421-1429.
- PR Newswire. Subtle Medical awarded breakthrough patent for reduced contrast agent dosage in medical imaging exams. May 4, 2021. https://www.prnewswire.com/news-releases/subtle-medical-awarded- breakthrough-patent-for-reduced-contrast-agent-dosage-in-medical- imaging-exams-301282780.html. Accessed March 17, 2022.