Ultra-low-field R<sub>1</sub> mapping at 0.2 mT obtained by field-cycling imaging correlates with cerebral small vessel disease burden.
Authors
Affiliations (5)
Affiliations (5)
- Aberdeen Biomedical Imaging Centre, Institute of Medical Sciences, The School of Medicine, Medical Sciences and Nutrition, University of Aberdeen, Scotland, AB25 2ZD, UK. [email protected].
- Aberdeen Biomedical Imaging Centre, Institute of Medical Sciences, The School of Medicine, Medical Sciences and Nutrition, University of Aberdeen, Scotland, AB25 2ZD, UK.
- Stavanger University Hospital, Stavanger, Norway.
- School of Natural and Computing Sciences, University of Aberdeen, Scotland, AB25 2ZD, UK.
- Institute of Medical Sciences, The School of Medicine, Medical Sciences and Nutrition, University of Aberdeen, Scotland, AB25 2ZD, UK.
Abstract
Field-cycling imaging (FCI) provides access to a unique regime of molecular dynamics across multiple ultra-low magnetic field strengths below 0.2 Tesla. This first proof-of-concept study aimed to determine whether ultra-low-field R<sub>1</sub> mapping obtained using FCI is sensitive to the presence and burden of cerebral small vessel disease. FCI was performed on 39 participants at field strengths of 0.2, 2, 20 and 200 mTesla, acquiring a single image slice. Sensitivity and specificity to discriminate the presence or absence of moderate or severe SVD by 1) two independent rater examinations of R<sub>1</sub> maps, 2) R<sub>1</sub> histogram features, and 3) deep learning classification were examined. Radiological review of co-registered 3 T MRI FLAIR volumes resulted in a final split cohort of 16:21 participants between a disease group and control group identified with and without moderate or severe WMHs, respectively. From FCI, the two groups were discriminated by Rater 1, Rater 2, best histogram feature, and best deep learning classification with sensitivity and specificity values of (0.81; 0.86), (0.88; 0.52), (0.94; 0.90), and (0.67; 1.00), respectively. Quantitative ultra-low multi-field R<sub>1</sub> mapping performed using FCI has diagnostic potential to detect brain tissue changes arising from cerebral small vessel disease.