He Wang1, Jinping Sun2, Dong Cui1, Xin Wang1, Jingna Jin1, Ying Li1, Zhipeng Liu1, Tao Yin1,3. 1. Institute of Biomedical Engineering, Chinese Academy of Medical Science & Peking Union Medical College, Tianjin, China. 2. School of Biomedical Engineering and Technology, Tianjin Medical University, Tianjin, China. 3. Neuroscience Center, Chinese Academy of Medical Science & Peking Union Medical College, Beijing, China.
Abstract
BACKGROUND: Inter-individual variability is an inherent and ineradicable feature of group-level brain atlases that undermines their reliability for clinical and other applications. To date, there have been no reports quantifying inter-individual variability in brain atlases. METHODS: In the present study, we compared inter-individual variability in nine brain atlases by task-based functional magnetic resonance imaging (MRI) mapping of motor and temporal lobe language regions in both cerebral hemispheres. We analyzed complete motor and language task-based fMRI and T1 data for 893 young, healthy subjects in the Human Connectome Project database. Euclidean distances (EDs) between hotspots in specific brain regions were calculated from task-based fMRI and brain atlas data. General linear model parameters were used to investigate the influence of different brain atlases on signal extraction. Finally, the inter-individual variability of ED and extracted signals and interdependence of relevant indicators were statistically evaluated. RESULTS: We found that inter-individual variability of ED varied across the nine brain atlases (P<0.0001 for motor regions and P<0.0001 for language regions). There was no correlation between parcel number and inter-individual variability in left to right (LtoR; P=0.7959 for motor regions and P=0.2002 for language regions) and right to left (RtoL; P=0.7654 for motor regions and P=0.3544 for language regions) ED; however, LtoR (P≤0.0001) and RtoL (P≤0.0001) inter-individual variability differed according to brain region: the LtoR (P=0.0008) and RtoL (P=0.0004) inter-individual variability was greater for the right hand than for the left hand, the LtoR (P=0.0019) and RtoL (P=0.0179) inter-individual variability was greater for the right language than for the left language, but there was no such difference between the right foot and left foot (LtoR, P=0.2469 and RtoL, P=0.6140). Inter-individual variability in one motor region was positively correlated with mean values in the other three motor regions (left hand, P=0.0145; left foot, P=0.0103; right hand, P=0.1318; right foot, P=0.3785). Inter-individual variability in language region was positively correlated with mean values in the four motor regions (left language, P=0.0422; right language, P=0.0514). Signal extraction for LtoR (P<0.0001) and RtoL (P<0.0001) varied across the nine brain atlases, which also showed differences in inter-individual variability. CONCLUSIONS: These results underscore the importance of quantitatively assessing the inter-individual variability of a brain atlas prior to use, and demonstrate that mapping motor regions by task-based fMRI is an effective method for quantitatively assessing the inter-individual variability in a brain atlas. 2021 Quantitative Imaging in Medicine and Surgery. All rights reserved.
BACKGROUND: Inter-individual variability is an inherent and ineradicable feature of group-level brain atlases that undermines their reliability for clinical and other applications. To date, there have been no reports quantifying inter-individual variability in brain atlases. METHODS: In the present study, we compared inter-individual variability in nine brain atlases by task-based functional magnetic resonance imaging (MRI) mapping of motor and temporal lobe language regions in both cerebral hemispheres. We analyzed complete motor and language task-based fMRI and T1 data for 893 young, healthy subjects in the Human Connectome Project database. Euclidean distances (EDs) between hotspots in specific brain regions were calculated from task-based fMRI and brain atlas data. General linear model parameters were used to investigate the influence of different brain atlases on signal extraction. Finally, the inter-individual variability of ED and extracted signals and interdependence of relevant indicators were statistically evaluated. RESULTS: We found that inter-individual variability of ED varied across the nine brain atlases (P<0.0001 for motor regions and P<0.0001 for language regions). There was no correlation between parcel number and inter-individual variability in left to right (LtoR; P=0.7959 for motor regions and P=0.2002 for language regions) and right to left (RtoL; P=0.7654 for motor regions and P=0.3544 for language regions) ED; however, LtoR (P≤0.0001) and RtoL (P≤0.0001) inter-individual variability differed according to brain region: the LtoR (P=0.0008) and RtoL (P=0.0004) inter-individual variability was greater for the right hand than for the left hand, the LtoR (P=0.0019) and RtoL (P=0.0179) inter-individual variability was greater for the right language than for the left language, but there was no such difference between the right foot and left foot (LtoR, P=0.2469 and RtoL, P=0.6140). Inter-individual variability in one motor region was positively correlated with mean values in the other three motor regions (left hand, P=0.0145; left foot, P=0.0103; right hand, P=0.1318; right foot, P=0.3785). Inter-individual variability in language region was positively correlated with mean values in the four motor regions (left language, P=0.0422; right language, P=0.0514). Signal extraction for LtoR (P<0.0001) and RtoL (P<0.0001) varied across the nine brain atlases, which also showed differences in inter-individual variability. CONCLUSIONS: These results underscore the importance of quantitatively assessing the inter-individual variability of a brain atlas prior to use, and demonstrate that mapping motor regions by task-based fMRI is an effective method for quantitatively assessing the inter-individual variability in a brain atlas. 2021 Quantitative Imaging in Medicine and Surgery. All rights reserved.
Authors: N Tzourio-Mazoyer; B Landeau; D Papathanassiou; F Crivello; O Etard; N Delcroix; B Mazoyer; M Joliot Journal: Neuroimage Date: 2002-01 Impact factor: 6.556
Authors: Jonathan D Power; Alexander L Cohen; Steven M Nelson; Gagan S Wig; Kelly Anne Barnes; Jessica A Church; Alecia C Vogel; Timothy O Laumann; Fran M Miezin; Bradley L Schlaggar; Steven E Petersen Journal: Neuron Date: 2011-11-17 Impact factor: 17.173
Authors: Alexander G Huth; Wendy A de Heer; Thomas L Griffiths; Frédéric E Theunissen; Jack L Gallant Journal: Nature Date: 2016-04-28 Impact factor: 49.962