1694
Educational Course - Half Day (4 hours)
As the premier imaging research tool, MRI methods are of broad and current interest in research. However, MRI scans are not as widely used as they could be in the clinical practice with patients, including neglected and vulnerable populations (e.g., patients from rural and remote areas, patients from low and middle-income countries or from impoverished areas in high income countries). According to healthcare professionals, there are several impediments for high strength MRI usage, including the availability of the scanner, safety concerns (e.g., stringent criteria for ferromagnetic materials), and cost (which also limits the use of MRI for routine screening even where it is available). Ultra-low field MRI may provide a solution for these various issues. Firstly, it provides a cost-effective option making it more accessible for healthcare facilities. Secondly, ultra-low field MRI provides a more compact and portable system in comparison to current clinical MRI scanners. Thirdly, lower magnetic field strengths can be safer for certain patients (reducing the risk of adverse effects), and it does not require restricted access or specifically designed shielded rooms. Despite lower resolution of ultra-low field compared to high-field MRI, novel machine learning techniques may improve the image quality of the ultra-low field MRI data. In the present educational session, we will cover a breadth of topics in ultra-low field imaging, including: (i) an overview of ultra-low field MRI technology and AI approaches to maximize the quality (including SNR) of ultra-low field images (A/Prof Chen); (ii) a novel machine learning technique to generate synthetic high-resolution images from the ultra-low field images (Dr Islam); (iii) diffusion and connectomics at ultra-low field MRI (Prof Jones); (iv) an overview of application of ultra-low field MRI in patients in the acute setting (Prof Parizel, Prof Law) and chronic stage (Dr Dominguez, Prof Caeyenberghs); and (v) applications in low-income countries (Prof Deoni). Attendees will learn how ultra-low field MRI systems can not only revolutionise the way in which people receive medical treatment but also increase the opportunity to address the healthcare inequality for the rural, remote and low-income areas.
Upon completion of our educational session, participants should be able to:
• Understand the unique opportunities and challenges associated with conducting MRI experiments at ultra-low magnetic field strengths;
• Explore different post-processing strategies to maximize SNR for ultra-low field images;
• Know the current state-of-the-art clinical imaging applications in ultra-low field MRI, including ultra-low field MRI studies in patients with brain injury from neglected and vulnerable populations.
These objectives aim to provide a comprehensive understanding of the principles, challenges, and applications of ultra-low field MRI.
This ultra-low field MRI educational session is targeted primarily at neuroscientists, imaging specialists and clinicians interested in the unique opportunities that could be afforded by a ultra-low field MRI system.
Presentations
While ultra-low field MRI has lower resolution and signal-to-noise ratio compared to high-field MRI, novel artificial intelligence techniques are being developed that may improve the image quality of the ultra-low field MRI data. In this session, we present results of a study where we used an image-to-image translation deep learning model to improve the quality of ultra-low-field (64mT) MRI scans to generate synthetic high-field (3T) MRI scans in a group of patients with acquired brain injury. This proof-of-concept study offers valuable insights into structural changes in the brain, potentially aiding in lesion identification and in the diagnosis and management of patients with brain injuries.
Presenter
Juan Dominguez, Deakin University Melbourne, Victoria
Australia
The development of portable ultra-low field strength brain MRI is a breakthrough, with the potential to bring imaging to the patient, rather than the other way around. This opens new horizons for greater MRI accessibility, including bedside neuroimaging. Yet, the technology continues to elicit scepticism and doubts from the clinical community. The ‘killer application’ is yet to be defined. The aim of this presentation is to focus on current clinical results and future potential of the technology.
Presenter
Paul Parizel, University of Western Australia Perth, West-Australia
Australia
Portable MRI scanners, while often low-cost, face challenges like low SNR, gradient and field non-uniformity, and heating-induced drift. By addressing these issues we’ve advanced white matter imaging from ADC maps to DTI, spherical deconvolution, and whole-brain tractography, & tentatively explored connectomics and quantitative myelin metrics for edge-weights. While such advances democratize neuroimaging & expand research to diverse populations, low system cost is crucial for true accessibility.
Presenter
Derek Jones, Cardiff University Cardiff
United Kingdom
Low and ultra-low field MRI has the potential to reshape health equity - enabling medical imaging in settings that traditionally lack access. However, an inexpensive MRI system will, by itself, not improve access without concurrent innovation in personnel training (inc. maintenance technicians, radiographers, and radiologists), data connectivity, electricity access, and tailoring for health conditions that are more prevalent in low and middle-income countries.
Presenter
Sean Deoni, Bill and Melinda Gates Foundation Washington, Seatle
United States
This presentation provides an overview of ultra-low field MRI technology, focusing on its recent advancements and the critical role of computing algorithms. Advanced algorithms enhance data acquisition, signal processing, and image reconstruction, with AI and machine learning improving signal-to-noise ratio, denoising, and producing high-fidelity images. The presentation also highlights image analysis for clinical applications and optimized workflows, demonstrating the effectiveness of ultra-low field MRI in both clinical and research settings.
Presenter
Zhaolin Chen, Monash University Clayton, Victoria
Australia
The significant potential of ultra-low field (64mT) MRI for point-of-care imaging, particularly in resource-limited settings, is highlighted. However, its adoption has been limited by lower image quality compared to high-field MRI. In this presentation, cutting-edge generative AI techniques are explored to enhance image quality, thereby bridging the gap between ultra-low and high-field MRI for improved diagnostic accuracy. Challenges, methodologies, and outcomes of using advanced generative models are discussed, with their implications for clinical workflows and accessibility emphasized. Examples of AI-driven image reconstruction and enhancement are provided, focusing on the translational potential of these advancements in clinical and research applications.
Presenter
Tohid Islam, Monash University Clayton, Victoria
Australia