Affiliation:
1. Thompson Institute University of the Sunshine Coast Sunshine Coast Queensland Australia
2. Tri‐Institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS) Georgia State University, Georgia Institute of Technology, Emory University Atlanta Georgia USA
3. Thompson Brain and Mind Healthcare Birtinya Queensland Australia
Abstract
BackgroundThe brainstem is a crucial component of the central autonomic nervous (CAN) system. Functional MRI (fMRI) of the brainstem remains challenging due to a range of factors, including diverse imaging protocols, analysis, and interpretation.PurposeTo develop an fMRI protocol for establishing a functional atlas in the brainstem.Study TypeProspective cross‐sectional study.SubjectsTen healthy subjects (four males, six females).Field Strength/SequenceUsing a 3.0 Tesla MR scanner, we acquired T1‐weighted images and three different fMRI scans using fMRI protocols of the optimized functional Imaging of Brainstem (FIBS), the Human Connectome Project (HCP), and the Adolescent Brain Cognitive Development (ABCD) project.AssessmentThe temporal signal‐to‐noise‐ratio (TSNR) of fMRI data was compared between the FIBS, HCP, and ABCD protocols. Additionally, the main normalization algorithms (i.e., FSL‐FNIRT, SPM‐DARTEL, and ANTS‐SyN) were compared to identify the best approach to normalize brainstem data using root‐mean‐square (RMS) error computed based on manually defined reference points. Finally, a functional autonomic brainstem atlas that maps brainstem regions involved in the CAN system was defined using meta‐analysis and data‐driven approaches.Statistical TestsANOVA was used to compare the performance of different imaging and preprocessing pipelines with multiple comparison corrections (P ≤ 0.05). Dice coefficient estimated ROI overlap, with 50% overlap between ROIs identified in each approach considered significant.ResultsThe optimized FIBS protocol showed significantly higher brainstem TSNR than the HCP and ABCD protocols (P ≤ 0.05). Furthermore, FSL‐FNIRT RMS error (2.1 ± 1.22 mm; P ≤ 0.001) exceeded SPM (1.5 ± 0.75 mm; P ≤ 0.01) and ANTs (1.1 ± 0.54 mm). Finally, a set of 12 final brainstem ROIs with dice coefficient ≥0.50, as a step toward the development of a functional brainstem atlas.Data ConclusionThe FIBS protocol yielded more robust brainstem CAN results and outperformed both the HCP and ABCD protocols.Evidence Level2Technical EfficacyStage 1
Funder
National Health and Medical Research Council
Cited by
5 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献