20th Edition | ARIA-E, Synthetic MRI, and New Modalities!

Campbell Arnold
Jul 29
6 min read

“There are few things people fear more than cancer, with the possible exception of neurodegenerative diseases."

— Robb Wolf

Welcome to Radiology Access! your biweekly newsletter on the people, research, and technology transforming global imaging access.

In this issue, we cover:

Bringing ARIA-E Detection to the Point of Care
Prospective Study Shows Synthetic STIR Saves Time and Boosts Image Quality
EMVision’s emu Clinical Trial Takes Flight
Low-field Super-Resolution Challenge Enters Validation Phase

If you want to stay up-to-date with the latest in Radiology and AI, then don't forget to subscribe!

Bringing ARIA-E Detection to the Point of Care

How low-field scanners could bring side-effect monitoring to where treatments are delivered.

Photo Credit: Barakos et al., The Journal of Prevention of Alzheimer's Disease 2022 — *Photo Credit:* *Barakos et al., The Journal of Prevention of Alzheimer's Disease 2022*

As amyloid-targeting Alzheimer’s therapies gain traction, monitoring for side effects like amyloid-related imaging abnormalities with edema (ARIA-E) is becoming a critical part of patient care. Patients on these therapies often require repeated brain MRI scans—typically on expensive high-field systems—which drives up treatment costs and complexity. As the eligible patient population grows, access to timely imaging may become a major bottleneck, particularly in rural or underserved areas. This is where portable, lower-cost MRI could offer a meaningful solution.

At the 2025 Alzheimer’s Association International Conference, researchers from Washington University in St. Louis presented preliminary findings from an ongoing study evaluating Hyperfine’s 64mT portable MRI system for ARIA-E detection. Among 31 patients receiving anti-amyloid therapy, the device demonstrated 100% sensitivity for ARIA-E detection when compared to conventional MRI.

While further validation is needed, these early results suggest that portable MRI may offer a practical tool for expanding ARIA-E surveillance, especially in settings without easy access to high-field scanners. The study’s principal investigator, Dr. Tammie Benzinger, noted that, “This research could help alleviate the burden on families and facilities and improve overall access to care.” She expressed excitement at the possibility of placing portable MRI units directly in infusion clinics, where they could streamline monitoring. As Alzheimer’s care continues to scale globally, tools like Hyperfine’s Swoop may help close imaging gaps and bring follow-up care closer to the patients.

Bottom line: Early results suggest portable MRI can reliably detect ARIA-E in Alzheimer’s patients on anti-amyloid therapies, offering a scalable, lower-cost solution for monitoring.

Prospective Study Shows Synthetic STIR Saves Time and Boosts Image Quality

Could image synthesis be the next forefront for MR acceleration?

*Photo Credit:Tanenbaum et al., AJNR 2023*

STIR is a core MRI sequence for detecting soft tissue abnormalities and edema, but it’s time-consuming and often produces noisy images. To address this, Subtle Medical (full disclosure: my employer) developed SubtleSYNTH, an AI-powered algorithm that synthesizes spinal STIR images from standard T1- and T2-weighted sequences—providing the option to forgo a dedicated STIR scan. A recent European Radiology publication presents a prospective clinical evaluation of the technology, comparing synthetic STIR (Synth-STIR) to conventional STIR in a real-world setting.

The single-center study enrolled 199 patients and included both quantitative image analysis and blinded review by four radiologists. Key findings include:

~90 seconds saved per exam, with no STIR sequence acquired.
Better signal and contrast, with Synth-STIR showing significant improvements in signal-to-noise ratio and contrast-to-noise ratio in vertebrae and discs (p < 0.001).
Radiologist ratings favored Synth-STIR for noise level, sharpness, lower artifacts, and diagnostic confidence.
Diagnostic performance was equivalent across 14 common spinal findings, including bone marrow edema, disc pathology, and vertebral fractures.

The results highlight how synthetic MRI can improve efficiency without compromising clinical accuracy—particularly in the spine, where STIR is essential but often a bottleneck. With faster protocols, better image quality, and seamless integration into existing workflows, Synth-STIR represents a meaningful step toward scalable, AI-enhanced imaging. While MR acceleration through enhancement has become commonplace, synthetic sequences are just beginning to move from research to routine clinical use, opening up new possibilities for faster protocols.

Bottom line: Synth-STIR delivers faster scans and cleaner images without sacrificing diagnostic performance—bringing synthetic MRI one step closer to clinical routine.

EMVision’s emu Clinical Trial Takes Flight

And how it could lead to a new class of electromagnetic imaging devices.

Photo Credit: Angie Lee — *Photo Credit:* *Angie Lee*

EMVision has announced a major milestone in the clinical trial of its portable electromagnetic brain imaging device, emu. Patient recruitment is now underway at five of six clinical sites, with final IRB approval secured at the sixth. The trial spans leading comprehensive stroke centers, including University of Texas Health, Mayo Clinic Florida, and Mount Sinai in the US, as well as Royal Melbourne Hospital and Liverpool Hospital in Australia.

The company aims to enroll 300 suspected stroke patients over 6–12 months to assess emu’s ability to detect hemorrhage with >80% sensitivity and specificity—a critical step toward their goal of FDA De Novo clearance.

Unlike CT, emu uses non-ionizing radiofrequency waves to detect tissue changes. Additionally, compared to conventional MRI, electromagnetic imaging has much lower infrastructure requirements, enabling portability. While EMVision must pursue de novo clearance for emu, they also plan to leverage it for accelerated FDA 510(k) clearance for their next First Responder device.

This technological advance could be a game-changer in emergency, pre-hospital, or resource-limited settings— offering clinicians rapid insight into brain injury at the point of care. With device deployments and regulatory green lights now in place, the study moves into full swing, bringing EMVision one step closer to delivering a new diagnostic modality for acute brain conditions.

Bottom line: With recruitment now underway, EMVision’s portable electromagnetic scanner trial could mark a turning point in point-of-care imaging—bringing a new modality into the game.

Low-field Super-Resolution Challenge Enters Validation Phase

Are you ready to put your algorithm to the test?

Photo Credit: Lucas et al., Radiology 2025 — *Photo Credit:* *Lucas et al., Radiology 2025*

The MICCAI 2025 ULF-EnC Challenge—Enhancing Ultra-Low-Field MRI with Paired High-Field MRI Comparisons—has officially entered the validation phase, and registration is still open! This unique competition invites participants to develop methods for enhancing diagnostic image quality using paired 64mT ULF and 3T high-field brain MRIs (T1, T2, FLAIR). If you're interested in improving low-field image quality, now’s the time to dive in. You can access the dataset on the challenge website and help push the boundaries of accessible neuroimaging!

Bottom line: Don’t delay, sign up for the low-field MRI challenge today!

Resource Highlight: Call for Open Source Low-Field Papers

Photo Credit: Sarty et al., Frontiers in Neuroimaging 2025 — *Photo Credit:* *Sarty et al., Frontiers in Neuroimaging 2025*

Are you building or working with open-source low-field MRI? This is your chance to showcase your work in a special NMR in Biomedicine issue dedicated to advancing open-access, low-field MRI systems! This edition focuses on both hardware and software open source designs and is spearheaded by Sairam Geethanath at Johns Hopkins University in collaboration with Jason Stockmann and Lukas Winter. The issue is accepting submissions through March 15th, 2026 so you’ve got some time to cook.

Feedback

We’re eager to hear your thoughts as we continue to refine and improve RadAccess. Is there an article you expected to see but didn’t? Have suggestions for making the newsletter even better? Let us know! Reach out via email, LinkedIn, or X—we’d love to hear from you.

References

Disclaimer: There are no paid sponsors of this content. The opinions expressed are solely those of the newsletter authors, and do not necessarily reflect those of referenced works or companies.