Cape Town - 2026 ISMRM-ISMRT Annual Meeting and Exhibition • 09-14 May 2026
| 16:00 |
|
606-03-001.
Antenatal Maternal Anaemia and Infant Brain Structure: High (3T) and Ultra-Low-Field (64mT) MRI Findings from South Africa
Impact: This study demonstrates that the impact of maternal anaemia on child brain structure is detectable as early as infancy, with effects emerging at 1 year of age. Findings inform the feasibility of ULF MRI and need for optimising anaemia interventions.
|
|
| 16:11 |
|
606-03-002.
Enhanced Ultra Low-Field Diffusion Tensor Imaging with Direction-Dependent Bias Correction and Spatio-Angular Superresolution
Impact: Post-processing correction and recovery methods to enhance low-field
diffusion tensor imaging sequences can facilitate accurate mapping of white
matter microstructure in the human brain via portable MRI.
|
|
| 16:22 |
|
606-03-003.
Ultra-Low-Field NMR Relaxometry Reveals Anti-Invasive Effects of Bumetanide in Glioblastoma mouse model.
Impact:
The detection of bumetanide treatment effects using FFC-NMR relaxometry demonstrates its potential as a therapy for brain glioma. The contrast mechanism shown here also makes it possible to monitor the treatment non-invasively using Field-Cycling Imaging or low-field MRI systems. |
|
| 16:33 |
|
606-03-004.
Infant Brain Growth from Portable uLF MRI in LMICs: Site, Sex, and Nutrition Effects in the First 1,000 Days
Impact: By enabling
population-based infant brain growth charts from portable ultra-low-field MRI
across LMIC sites, this work supports earlier, equitable neurodevelopmental
surveillance and evaluation of nutrition and perinatal programs, guiding
evaluation of interventions when high-field MRI is unavailable or impractical.
|
|
| 16:44 |
|
606-03-005.
Generalization of low-field 3D MRI acceleration via the CIRIM network across knee, spine and brain
Impact: CIRIM deep learning reconstruction enables substantial scan time reduction in low-field MRI. By successfully accelerating 3D knee and spine imaging and generalizing across anatomies and fieldstrength, CIRIM demonstrates robust performance, advancing the practical use of accelerated reconstruction in low-field MRI.
|
|
| 16:55 |
|
606-03-006.
AUTOSEQ-based Magnetic Resonance Fingerprinting at 6.5 mT: A Framework for Low-Field Quantitative Breast Imaging
Impact: AUTOSEQ-based MR Fingerprinting at 6.5 mT provides accurate, reproducible T1/T2 measurements in breast-mimicking phantoms, validating the feasibility of quantitative relaxation measurements at ultra-low field and establishing a foundation for spatially resolved, low-cost breast imaging for point-of-care and early detection applications.
|
|
| 17:06 |
|
606-03-007.
Ultra-low-field Lumbar Spine MRI at 0.05 Tesla
Impact: This study demonstrates high-quality lumbar spine MRI on a 0.05 Tesla
whole-body MRI scanner by leveraging advanced computational modelling and
extensive high-field MRI data. These developments will facilitate a new class
of affordable, patient-centric, and computing-powered MRI scanners.
|
|
| 17:17 |
|
606-03-008.
SuperField-Net2: Simultaneous T1w-T2w MRI Enhancement from T2w Ultra-Low-Field Imaging via Frequency-Attenuation
Impact: SFNet2 enhances a single
T2w uLF MRI scan into HF-like paired T1w and T2w images, improving image
quality, resolution, and utility. SFNet2 could shorten MRI scan time, enable multi-sequence
imaging on portable scanners, and expand point-of-care neuroimaging in resource-limited
settings.
|
|
| 17:28 |
|
606-03-009.
The (Statistical) Power of Low Field MRI: Reliability and Repeatability of Diffusion MRI at 64 mT
Impact: Sample size calculations show that portable ultra-low-field diffusion MRI can enable population neuroscience and epidemiological studies, extending advanced neuroimaging to underserved populations beyond the reach of conventional high-field scanners.
|
|
| 17:39 |
|
606-03-010.
Low-field knee MRI in the clinical setting: a comparative study of a 72 mT prototype and a clinical 3T scanner
Impact: This work presents a paired low- and high-field knee MRI study, creating a clinical dataset for comparison and for developing deep-learning-based image enhancement methods. Common knee lesions were identifiable by radiologists in an initial evaluation.
|
© 2026 International Society for Magnetic Resonance in Medicine