Rachana R Borkar, Sai Dhanush Reddy Jeggari, Kamal Kandel, Kaviya Partheepan, Nishant Sharma, Sandeep Samethadka Nayak
Low-field and portable MRI show promising diagnostic potential for AIS and TIA, particularly when conventional MRI is unavailable, delayed, or impractical. However, current evidence is limited by small, predominantly single-center studies with substantial risk of bias, and further prospective multicenter validation is required before these technologies can be incorporated into routine clinical decision-making.
BACKGROUND: Magnetic resonance imaging (MRI) has a central role in acute ischemic stroke (AIS) and transient ischemic attack (TIA) diagnosis; however, conventional high-field MRI remains limited by infrastructure requirements, patient transport, and restricted accessibility. Low-field and portable MRI systems have emerged as potential solutions for point-of-care neuroimaging in emergency, intensive care, and resource-limited settings.
METHODS: A systematic review was conducted according to PRISMA 2020 guidelines. PubMed, Scopus, Web of Science, and Cochrane Library databases were searched from inception through May 2026. Studies evaluating low-field or portable MRI systems (≤0.55 T) in adults with AIS, TIA, sub-acute ischemic stroke, or suspected stroke were included. Diagnostic accuracy, feasibility, safety, workflow, and clinical utility outcomes were extracted. Risk of bias was assessed using QUADAS-2.
RESULTS: Eleven studies encompassing portable and low-field MRI platforms ranging from 0.064 T to 0.55 T were included. Portable MRI demonstrated feasibility in bedside ICU and emergency department settings without major device-related adverse events. Diagnostic performance varied by field strength, lesion size, and imaging protocol.
CONCLUSIONS: Low-field and portable MRI show promising diagnostic potential for AIS and TIA, particularly when conventional MRI is unavailable, delayed, or impractical. However, current evidence is limited by small, predominantly single-center studies with substantial risk of bias, and further prospective multicenter validation is required before these technologies can be incorporated into routine clinical decision-making.