Looping Star fMRI in Cognitive Tasks and Resting State. Issue 3 (19th February 2020)

• Record Type: Journal Article
• Authors: Dionisio‐Parra, Beatriz; Wiesinger, Florian; Sämann, Philipp G.; Czisch, Michael; Solana, Ana Beatriz
• Is Part Of: Journal of magnetic resonance imaging. Volume 52:Issue 3(2020)
• Page Start: 739
• Page End: 751
• Publication Date: 2020-02-19
• Subjects: fMRI -- quiet -- acoustic noise -- looping star -- Zero TE -- 3D radial -- RUFIS -- silent; Magnetic resonance imaging -- Periodicals; 616
• DOI: 10.1002/jmri.27073
• Languages: English
• ISSNs: 1053-1807

Abstract

Abstract : Background: Conventional T2 *‐weighted functional magnetic resonance imaging (fMRI) is performed with echo‐planar imaging (EPI) sequences that create substantial acoustic noise. The loud acoustic noise not only affects the activation of the auditory cortex, but may also interfere with resting state and task fMRI experiments. Purpose: To demonstrate the feasibility of a novel, quiet, T2 *, whole‐brain blood oxygenation level‐dependent (BOLD)‐fMRI method, termed Looping Star, compared to conventional multislice gradient‐echo EPI. Study Type: Prospective. Phantom/Subjects: Glover stability QA phantom; 10 healthy volunteers. Field Strength/Sequence: 3.0T: gradient echo (GE)‐EPI and T2 * Looping Star fMRI. Assessment: Looping Star fMRI was presented and compared to GE‐EPI with a working memory (WM) task and resting state (RS) experiments. Temporal stability and acoustic measurements were obtained for both methods. Functional maps and activation accuracy were compared to evaluate the performance of the novel sequence. Statistical Tests: Mean and standard deviation values were analyzed for temporal stability and acoustic noise tests. Activation maps were assessed with one‐sample t ‐tests and contrast estimates (CE). Paired t ‐tests and receiver operator characteristic (ROC) were used to compare fMRI sensitivity and performance. Results: Looping Star presented a 98% reduction in sound pressure compared with GE‐EPI, with stable temporal stability (0.09% percent fluctuation), but reduced temporal signal‐to‐noise ratio (tSNR) (mean difference = 15.9%). The novel method yielded consistent activations for RS and WM (83.4% and 69.5% relative BOLD sensitivity), which increased with task difficulty (mean CE 2‐back = 0.56 vs. 0‐back = 0.08, P < 0.05). A few differences in spatial activations were found between sequences, leading to a 4–8% lower activation accuracy with Looping Star. Data Conclusion: Looping Star provides a suitable approach for whole‐brain coverage with sufficient spatiotemporal resolution and BOLD sensitivity, with only 0.5 dB above ambient noise. From the comparison with GE‐EPI, further developments of Looping Star fMRI should target increased sensitivity and spatial specificity for both RS and task experiments. Level of Evidence: 2. Technical Efficacy Stage: 1 J. Magn. Reson. Imaging 2020;52:739–751.