Skip to main content
2024 AANS Annual Scientific Meeting Main banner
Back

Neuro Spot Light Oral

Neuro Spotlight

Exploring Ultra-Low-Field Portable Magnetic Resonance Imaging (ULF-pMRI) for Brain Tumor Assessment in LMIC: A Preliminary Study

Saturday, May 4, 2024
2:36 PM – 2:38 PM CT
Location:  Booth 745 Neuro Spotlight
Disclosure(s):
Ahmed altaf, MD: No financial relationships to disclose
Syed Ather Enam, MD, Phd: No financial relationships to disclose
Introduction: In low- and middle-income countries (LMICs), limited access to neuroimaging with MRI complicates brain tumor diagnosis and management. Conventional high-field MRI (cMRI) systems are expensive, unavailable in many LMIC healthcare facilities. Ultra-low-field portable MRI (ULF pMRI) is a promising, lower-cost alternative that improves accessibility to brain imaging in resource-constrained settings. Our objective was to determine if ULF pMRI enables adequate visualization and quantification of brain tumor morphology for guiding neurosurgical decisions without cMRI access.

Methods: Patients with suspected brain tumors who underwent neuroimaging from August to November 2022 were enrolled in this study. Using a 0.06T ULF pMRI scanner (Hyperfine Swoop®) and a 1.5T cMRI system (General Electric Optima MR450w). The same imaging sequences were acquired on both scanners including T1W, T2W, FLAIR. pMRI scanning was conducted at the patient's bedside with a total protocol time of under 45 minutes. Brain tumor volumes were segmented manually on T2 and T1-weighted images and measured by a radiologist blinded to MRI type. Volumetrics were compared between pMRI and cMRI using Bland-Altman analysis and correlation coefficients to assess agreement.

Results: This study included 30 patients (mean age 41.3 ± 12.1 years) with brain tumors who underwent both ULF pMRI and cMRI. Comparative analysis revealed some variations in tumor dimensions between the two modalities among individual patients. However, a strong overall correlation was observed, with no clinically significant differences per Bland-Altman plots (r = 0.96, p < 0.001), with a mean difference of -1.2 cc (95% limits of agreement -3.4 to 0.9 cc). Importantly, ULF pMRI successfully delineated key features needed for surgical planning including tumor margins, edema, effects on the vasculature, and relationship with eloquent brain areas. Visualization of these characteristics was comparable to cMRI.

Conclusion : The strong correlation and negligible volumetric differences versus 1.5T MRI indicate reliable tumor delineation for surgical planning. Our findings suggest that ULF pMRI is a viable alternative to diagnosing brain tumors. ULF pMRI successfully characterized tumor morphology, edema, vasculature, and eloquent area relations critical for management.