Diagnostic Accuracy of Portable Low- and Ultra-Low-Field MRI for Acute Ischemic Stroke: A Systematic Review and Meta-Analysis
Year: 2026
Type: article
Study designs: Systematic Review, Meta-Analysis
Abstract: BACKGROUND: Portable low-field and ultra-low-field MRI enables bedside imaging, but its diagnostic performance for acute ischemic stroke (AIS) detection remains uncertain. PURPOSE: To evaluate the diagnostic performance of portable MRI for AIS detection. DATA SOURCES: Four databases were searched on February 2, 2026. STUDY SELECTION: Eligible studies enrolled adults with suspected or confirmed AIS undergoing portable MRI. Nine studies comprising 474 patients with AIS were included. DATA ANALYSIS: In studies with an appropriate reference standard, pooled sensitivity and specificity were synthesized using a bivariate random-effects model, and a hierarchical summary receiver operating characteristic curve was generated. Across all eligible AIS cohorts, the pooled lesion detection rate was estimated using a generalized linear mixed-effects model with random effects. DATA SYNTHESIS: In six comparative studies, the pooled sensitivity for AIS detection was 73.3% (95% CI, 65.4-80.0), corresponding to approximately one in four confirmed infarcts being missed, with missed lesions generally small or sub-centimetric. Pooled specificity was 79.3% (95% CI, 69.1-86.8), and the area under the curve was 0.818. The pooled AIS lesion detection rate was 87.65% (95% CI, 75.7-94.2). LIMITATIONS: = 74.2%). This estimate was interpreted as secondary and descriptive because it included case-only confirmed AIS cohorts. CONCLUSIONS: Portable low/ultra-low-field MRI showed moderate diagnostic performance for AIS detection in comparative studies and may serve as a bedside adjunct where conventional MRI is constrained. However, given the false-negative risk, portable MRI should not be relied upon as a stand-alone rule-out test for AIS when clinical suspicion persists.
DOI: 10.3174/ajnr.a9532
Language: English
Subfield: Epidemiology
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
OpenAlex ID: W7168465672