The Human Connectome Project's neuroimaging approach

Abstract

Noninvasive human neuroimaging has yielded many discoveries about the brain. Numerous methodological advances have also occurred, though inertia has slowed their adoption. This paper presents an integrated approach to data acquisition, analysis and sharing that builds upon recent advances, particularly from the Human Connectome Project (HCP). The 'HCP-style' paradigm has seven core tenets: (i) collect multimodal imaging data from many subjects; (ii) acquire data at high spatial and temporal resolution; (iii) preprocess data to minimize distortions, blurring and temporal artifacts; (iv) represent data using the natural geometry of cortical and subcortical structures; (v) accurately align corresponding brain areas across subjects and studies; (vi) analyze data using neurobiologically accurate brain parcellations; and (vii) share published data via user-friendly databases. We illustrate the HCP-style paradigm using existing HCP data sets and provide guidance for future research. Widespread adoption of this paradigm should accelerate progress in understanding the brain in health and disease.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Figure 1: Combined representation of cortical surface vertices and subcortical voxels in the CIFTI grayordinates standard space.
Figure 2: Improved intersubject registration using information based on areal features in addition to cortical folding.
Figure 3: The HCP_MMP1.0.
Figure 4: Example parcellated analyses run using HCP data and the HCP_MMP1.0 cortical parcellation.
Figure 5: Current and future projects that use HCP-style data acquisition, analysis and sharing.
Figure 6: Fidelity of localizing area MT+ (architectonic area hOc5) when mapped to the cortical surface by different methods.

References

  1. 1

    Behrens, T.E. & Sporns, O. Human connectomics. Curr. Opin. Neurobiol. 22, 144–153 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. 2

    Jbabdi, S., Sotiropoulos, S.N., Haber, S.N., Van Essen, D.C. & Behrens, T.E. Measuring macroscopic brain connections in vivo. Nat. Neurosci. 18, 1546–1555 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. 3

    Smith, S. Introduction to the NeuroImage special issue “Mapping the Connectome”. Neuroimage 80, 1 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  4. 4

    Van Essen, D.C. et al. The WU-Minn Human Connectome Project: an overview. Neuroimage 80, 62–79 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  5. 5

    Ugˇurbil, K. et al. Pushing spatial and temporal resolution for functional and diffusion MRI in the Human Connectome Project. Neuroimage 80, 80–104 (2013).

    Article  CAS  Google Scholar 

  6. 6

    Glasser, M.F. et al. The minimal preprocessing pipelines for the Human Connectome Project. Neuroimage 80, 105–124 (2013). This study provides an overview of the preprocessing pipelines used for the structural, functional and diffusion MRI modalities acquired by the HCP and by a growing number of HCP-style projects. It also introduces the concept of grayordinates, representing surface and volume data in a combined CIFTI data format.

    Article  PubMed  PubMed Central  Google Scholar 

  7. 7

    Griffanti, L. et al. ICA-based artefact removal and accelerated fMRI acquisition for improved resting state network imaging. Neuroimage 95, 232–247 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  8. 8

    Salimi-Khorshidi, G. et. al. Automatic denoising of functional MRI data: combining independent component analysis and hierarchical fusion of classifiers. Neuroimage 90, 449–468 (2014). This study demonstrated a method for automatically cleaning spatially specific structured noise out of fMRI data using an independent component analysis (ICA) decomposition and machine learning classification of the resulting components into signal and noise.

    Article  PubMed  PubMed Central  Google Scholar 

  9. 9

    Glasser, M.F. et al. A multi-modal parcellation of human cerebral cortex. Nature nature18933 (2016). This study describes the HCP_MMP1.0 multimodal human cortical parcellation in a group average and in individual subjects. Extensive supplemental data detail the analysis methods used and the evidence supporting the identification of all 180 areas in each hemisphere.

  10. 10

    Robinson, E.C. et al. MSM: a new flexible framework for multimodal surface matching. Neuroimage 100, 414–426 (2014). This study introduced a powerful new method for aligning cortical surfaces from individuals to an atlas using areal features as well as folding patterns to constrain the alignment.

    Article  PubMed  PubMed Central  Google Scholar 

  11. 11

    Van Essen, D.C. et al. The brain analysis of spatial maps and atlases (BALSA) database. Neuroimage j.neuroimage.2016.04.002 (2016).

  12. 12

    Marcus, D.S. et al. Human Connectome Project informatics: quality control, database services, and data visualization. Neuroimage 80, 202–219 (2013).

    Article  PubMed  Google Scholar 

  13. 13

    Eklund, A., Nichols, T.E. & Knutsson, H. Cluster failure: why fMRI inferences for spatial extent have inflated false-positive rates. Proc. Natl. Acad. Sci. USA 28, 7900–7905 (2016).

    Article  CAS  Google Scholar 

  14. 14

    Turner, R. Where matters: new approaches to brain analysis. in Microstructural Parcellation of the Human Cerebral Cortex (eds. Geyer, S. & Turner, R.) 179–196 (Springer, 2013).

  15. 15

    Turner, R. & Geyer, S. Comparing like with like: the power of knowing where you are. Brain Connect. 4, 547–557 (2014). The authors emphasize the importance of high-quality data acquisition, careful surface-based analysis in individual subjects and localization in relation to the best available cortical parcellation. Hence they presage many of the points made in the present article.

    Article  PubMed  PubMed Central  Google Scholar 

  16. 16

    Sereno, M.I. et al. Borders of multiple visual areas in humans revealed by functional magnetic resonance imaging. Science 268, 889–893 (1995).

    Article  CAS  Google Scholar 

  17. 17

    Wang, L., Mruczek, R.E., Arcaro, M.J. & Kastner, S. Probabilistic maps of visual topography in human cortex. Cereb. Cortex 25, 3911–3931 (2015).

    Article  CAS  PubMed  Google Scholar 

  18. 18

    Fischl, B. & Dale, A.M. Measuring the thickness of the human cerebral cortex from magnetic resonance images. Proc. Natl. Acad. Sci. USA 97, 11050–11055 (2000).

    Article  CAS  Google Scholar 

  19. 19

    Glasser, M.F., Goyal, M.S., Preuss, T.M., Raichle, M.E. & Van Essen, D.C. Trends and properties of human cerebral cortex: correlations with cortical myelin content. Neuroimage 93, 165–175 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. 20

    Glasser, M.F. & Van Essen, D.C. Mapping human cortical areas in vivo based on myelin content as revealed by T1- and T2-weighted MRI. J. Neurosci. 31, 11597–11616 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. 21

    Dale, A.M., Fischl, B. & Sereno, M.I. Cortical surface-based analysis. I. Segmentation and surface reconstruction. Neuroimage 9, 179–194 (1999).

    Article  CAS  Google Scholar 

  22. 22

    Dale, A.M. & Sereno, M.I. Improved localizadon of cortical activity by combining EEG and MEG with MRI cortical surface reconstruction: a linear approach. J. Cogn. Neurosci. 5, 162–176 (1993).

    Article  CAS  PubMed  Google Scholar 

  23. 23

    Fischl, B., Liu, A. & Dale, A.M. Automated manifold surgery: constructing geometrically accurate and topologically correct models of the human cerebral cortex. IEEE Trans. Med. Imaging 20, 70–80 (2001).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. 24

    Fischl, B. et al. Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron 33, 341–355 (2002).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. 25

    Fischl, B. et al. Sequence-independent segmentation of magnetic resonance images. Neuroimage 23 (Suppl. 1), S69–S84 (2004).

    Article  PubMed  PubMed Central  Google Scholar 

  26. 26

    Fischl, B., Sereno, M.I. & Dale, A.M. Cortical surface-based analysis. II: inflation, flattening, and a surface-based coordinate system. Neuroimage 9, 195–207 (1999).

    Article  CAS  Google Scholar 

  27. 27

    Fischl, B., Sereno, M.I., Tootell, R.B. & Dale, A.M. High-resolution intersubject averaging and a coordinate system for the cortical surface. Hum. Brain Mapp. 8, 272–284 (1999). This study introduced an automated method for aligning cortical surfaces from individuals to a surface-based atlas using cortical folding to constrain the registration.

    Article  CAS  PubMed  Google Scholar 

  28. 28

    Ségonne, F., Pacheco, J. & Fischl, B. Geometrically accurate topology-correction of cortical surfaces using nonseparating loops. IEEE Trans. Med. Imaging 26, 518–529 (2007).

    Article  PubMed  Google Scholar 

  29. 29

    Barch, D.M. et al. Function in the human connectome: task-fMRI and individual differences in behavior. Neuroimage 80, 169–189 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  30. 30

    Wandell, B.A. & Winawer, J. Imaging retinotopic maps in the human brain. Vision Res. 51, 718–737 (2011).

    Article  PubMed  PubMed Central  Google Scholar 

  31. 31

    Orban, G.A., Zhu, Q. & Vanduffel, W. The transition in the ventral stream from feature to real-world entity representations. Front. Psychol. 5, 695 (2014).

    PubMed  PubMed Central  Google Scholar 

  32. 32

    Brewer, A.A., Press, W.A., Logothetis, N.K. & Wandell, B.A. Visual areas in macaque cortex measured using functional magnetic resonance imaging. J. Neurosci. 22, 10416–10426 (2002).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. 33

    Van Essen, D.C., Felleman, D.J., DeYoe, E.A., Olavarria, J. & Knierim, J. Modular and hierarchical organization of extrastriate visual cortex in the macaque monkey. Cold Spring Harb. Symp. Quant. Biol. 55, 679–696 (1990).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. 34

    Calabrese, E., Badea, A., Cofer, G., Qi, Y. & Johnson, G.A. A diffusion MRI tractography connectome of the mouse brain and comparison with neuronal tracer data. Cereb. Cortex 25, 4628–4637 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  35. 35

    Donahue, C.J. et al. Using diffusion tractography to predict cortical connection strength and distance: A quantitative comparison with tracers in the monkey. J. Neurosci 36, 6758–6770 (2016).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. 36

    Basser, P.J., Mattiello, J. & LeBihan, D. MR diffusion tensor spectroscopy and imaging. Biophys. J. 66, 259–267 (1994).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. 37

    Fieremans, E., Jensen, J.H. & Helpern, J.A. White matter characterization with diffusional kurtosis imaging. Neuroimage 58, 177–188 (2011).

    Article  PubMed  PubMed Central  Google Scholar 

  38. 38

    Zhang, H., Schneider, T., Wheeler-Kingshott, C.A. & Alexander, D.C. NODDI: practical in vivo neurite orientation dispersion and density imaging of the human brain. Neuroimage 61, 1000–1016 (2012).

    Article  PubMed  PubMed Central  Google Scholar 

  39. 39

    Van Essen, D.C., Glasser, M.F., Dierker, D.L., Harwell, J. & Coalson, T. Parcellations and hemispheric asymmetries of human cerebral cortex analyzed on surface-based atlases. Cereb. Cortex 22, 2241–2262 (2012).

    Article  PubMed  PubMed Central  Google Scholar 

  40. 40

    Liewald, D., Miller, R., Logothetis, N., Wagner, H.J. & Schüz, A. Distribution of axon diameters in cortical white matter: an electron-microscopic study on three human brains and a macaque. Biol. Cybern. 108, 541–557 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  41. 41

    Van Essen, D.C. & Ugurbil, K. The future of the human connectome. Neuroimage 62, 1299–1310 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. 42

    Huang, S.Y. et al. The impact of gradient strength on in vivo diffusion MRI estimates of axon diameter. Neuroimage 106, 464–472 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  43. 43

    McNab, J.A. et al. The Human Connectome Project and beyond: initial applications of 300 mT/m gradients. Neuroimage 80, 234–245 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  44. 44

    Setsompop, K. et al. Blipped-controlled aliasing in parallel imaging for simultaneous multislice echo planar imaging with reduced g-factor penalty. Magn. Reson. Med. 67, 1210–1224 (2012). This study introduced a slice gradient blip technique for simultaneous multislice (multiband) echo-planar imaging to significantly reduce image artifacts in unaliasing of simultaneously acquired slices, particularly for high multiband factors. This technique is adopted in the HCP pulse sequences for fMRI and dMRI.

    Article  PubMed  PubMed Central  Google Scholar 

  45. 45

    Setsompop, K. et al. Pushing the limits of in vivo diffusion MRI for the Human Connectome Project. Neuroimage 80, 220–233 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. 46

    Sotiropoulos, S.N. et al. Fusion in diffusion MRI for improved fibre orientation estimation: An application to the 3T and 7T data of the Human Connectome Project. Neuroimage 134, 396–409 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

  47. 47

    Sotiropoulos, S.N. et al. Advances in diffusion MRI acquisition and processing in the Human Connectome Project. Neuroimage 80, 125–143 (2013). The authors report on advances in acquisition and analysis of HCP diffusion imaging data. This includes use of scanners with high maximal gradient strength, multiband pulse sequences, and improvements in image reconstruction, distortion correction, fiber orientation modeling and probabilistic tractography algorithms.

    Article  PubMed  PubMed Central  Google Scholar 

  48. 48

    Vu, A.T. et al. High resolution whole brain diffusion imaging at 7T for the Human Connectome Project. Neuroimage 122, 318–331 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. 49

    Hillman, E.M. Coupling mechanism and significance of the BOLD signal: a status report. Annu. Rev. Neurosci. 37, 161–181 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  50. 50

    Feinberg, D.A. et al. Multiplexed echo planar imaging for sub-second whole brain FMRI and fast diffusion imaging. PLoS One 5, e15710 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. 51

    Chen, J.E. & Glover, G.H. BOLD fractional contribution to resting-state functional connectivity above 0.1 Hz. Neuroimage 107, 207–218 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  52. 52

    Niazy, R.K., Xie, J., Miller, K., Beckmann, C.F. & Smith, S.M. Spectral characteristics of resting state networks. Prog. Brain Res. 193, 259–276 (2011).

    Article  PubMed  PubMed Central  Google Scholar 

  53. 53

    Auerbach, E.J., Xu, J., Yacoub, E., Moeller, S. & Ugˇurbil, K. Multiband accelerated spin-echo echo planar imaging with reduced peak RF power using time-shifted RF pulses. Magn. Reson. Med. 69, 1261–1267 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  54. 54

    Moeller, S. et al. Multiband multislice GE-EPI at 7 tesla, with 16-fold acceleration using partial parallel imaging with application to high spatial and temporal whole-brain fMRI. Magn. Reson. Med. 63, 1144–1153 (2010). This study demonstrated the feasibility of improving the spatial and temporal resolution of whole-brain EPI-based fMRI by slice multiplexing, in which multiple slices are excited and acquired simultaneously, thus sharing a common contrast and common spatial encoding periods. Multiband EPI is a key strength of the HCP fMRI and dMRI pulse sequences.

    Article  PubMed  PubMed Central  Google Scholar 

  55. 55

    Xu, J. et al. Evaluation of slice accelerations using multiband echo planar imaging at 3 T. Neuroimage 83, 991–1001 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  56. 56

    Laumann, T.O. et al. Functional system and areal organization of a highly sampled individual human brain. Neuron 87, 657–670 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. 57

    Jbabdi, S., Sotiropoulos, S.N., Savio, A.M., Graña, M. & Behrens, T.E. Model-based analysis of multishell diffusion MR data for tractography: how to get over fitting problems. Magn. Reson. Med. 68, 1846–1855 (2012).

    Article  PubMed  PubMed Central  Google Scholar 

  58. 58

    Jeurissen, B., Tournier, J.D., Dhollander, T., Connelly, A. & Sijbers, J. Multi-tissue constrained spherical deconvolution for improved analysis of multi-shell diffusion MRI data. Neuroimage 103, 411–426 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  59. 59

    Greve, D.N. & Fischl, B. Accurate and robust brain image alignment using boundary-based registration. Neuroimage 48, 63–72 (2009).

    Article  PubMed  PubMed Central  Google Scholar 

  60. 60

    Andersson, J.L. & Sotiropoulos, S.N. Non-parametric representation and prediction of single- and multi-shell diffusion-weighted MRI data using Gaussian processes. Neuroimage 122, 166–176 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  61. 61

    Andersson, J.L. & Sotiropoulos, S.N. An integrated approach to correction for off-resonance effects and subject movement in diffusion MR imaging. Neuroimage 125, 1063–1078 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

  62. 62

    Andersson, J.L.R., Graham, M.S., Zsoldos, E. & Sotiropoulos, S.N. Incorporating outlier detection and replacement into a non-parameteric framework for movement and distortion correction of diffusion MR images. Neuroimage j.neuroimage.2016.06.058 (2016).

  63. 63

    Power, J.D., Schlaggar, B.L. & Petersen, S.E. Recent progress and outstanding issues in motion correction in resting state fMRI. Neuroimage 105, 536–551 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  64. 64

    Power, J.D., Barnes, K.A., Snyder, A.Z., Schlaggar, B.L. & Petersen, S.E. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. Neuroimage 59, 2142–2154 (2012).

    Article  PubMed  PubMed Central  Google Scholar 

  65. 65

    Satterthwaite, T.D. et al. Impact of in-scanner head motion on multiple measures of functional connectivity: relevance for studies of neurodevelopment in youth. Neuroimage 60, 623–632 (2012).

    Article  PubMed  PubMed Central  Google Scholar 

  66. 66

    Beckmann, C.F. & Smith, S.M. Probabilistic independent component analysis for functional magnetic resonance imaging. IEEE Trans. Med. Imaging 23, 137–152 (2004).

    Article  PubMed  PubMed Central  Google Scholar 

  67. 67

    Winkler, A.M., Webster, M.A., Vidaurre, D., Nichols, T.E. & Smith, S.M. Multi-level block permutation. Neuroimage 123, 253–268 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  68. 68

    Zhang, H. et al. High-dimensional spatial normalization of diffusion tensor images improves the detection of white matter differences: an example study using amyotrophic lateral sclerosis. IEEE Trans. Med. Imaging 26, 1585–1597 (2007).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  69. 69

    Zhang, H., Yushkevich, P.A., Alexander, D.C. & Gee, J.C. Deformable registration of diffusion tensor MR images with explicit orientation optimization. Med. Image Anal. 10, 764–785 (2006).

    Article  PubMed  PubMed Central  Google Scholar 

  70. 70

    Amunts, K., Malikovic, A., Mohlberg, H., Schormann, T. & Zilles, K. Brodmann's areas 17 and 18 brought into stereotaxic space-where and how variable? Neuroimage 11, 66–84 (2000). This study (and many others from the same laboratory) used objective observer-independent methods to identify cytoarchitectonic areas in post-mortem human brains.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  71. 71

    Eickhoff, S.B. et al. A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data. Neuroimage 25, 1325–1335 (2005).

    Article  PubMed  PubMed Central  Google Scholar 

  72. 72

    Van Essen, D.C. A population-average, landmark- and surface-based (PALS) atlas of human cerebral cortex. Neuroimage 28, 635–662 (2005).

    Article  PubMed  PubMed Central  Google Scholar 

  73. 73

    Anticevic, A. et al. Comparing surface-based and volume-based analyses of functional neuroimaging data in patients with schizophrenia. Neuroimage 41, 835–848 (2008).

    Article  PubMed  PubMed Central  Google Scholar 

  74. 74

    Fischl, B. et al. Cortical folding patterns and predicting cytoarchitecture. Cereb. Cortex 18, 1973–1980 (2008). The authors used folding-based alignment to register cytoarchitectonic areas to a surface-based atlas. In the resultant probabilistic maps, alignment was high for early sensory areas but much lower for 'higher' cortical areas, reflecting regional differences in the correlation between areal boundaries and cortical folding.

    Article  PubMed  PubMed Central  Google Scholar 

  75. 75

    Frost, M.A. & Goebel, R. Measuring structural-functional correspondence: spatial variability of specialised brain regions after macro-anatomical alignment. Neuroimage 59, 1369–1381 (2012).

    Article  PubMed  PubMed Central  Google Scholar 

  76. 76

    Smith, S.M. et al. Resting-state fMRI in the Human Connectome Project. Neuroimage 80, 144–168 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  77. 77

    Tucholka, A., Fritsch, V., Poline, J.B. & Thirion, B. An empirical comparison of surface-based and volume-based group studies in neuroimaging. Neuroimage 63, 1443–1453 (2012).

    Article  PubMed  PubMed Central  Google Scholar 

  78. 78

    Felleman, D.J. & Van Essen, D.C. Distributed hierarchical processing in the primate cerebral cortex. Cereb. Cortex 1, 1–47 (1991).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  79. 79

    Haxby, J.V. et al. A common, high-dimensional model of the representational space in human ventral temporal cortex. Neuron 72, 404–416 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  80. 80

    Glasser, M.F. & Rilling, J.K. DTI tractography of the human brain's language pathways. Cereb. Cortex 18, 2471–2482 (2008).

    Article  PubMed  PubMed Central  Google Scholar 

  81. 81

    Brodmann, K. Vergleichende Lokalisationslehre der Grosshirnrinde (Leipzig, 1909).

  82. 82

    Nieuwenhuys, R. The myeloarchitectonic studies on the human cerebral cortex of the Vogt-Vogt school, and their significance for the interpretation of functional neuroimaging data. Brain Struct. Funct. 218, 303–352 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  83. 83

    Vogt, C. & Vogt, O. Allgemeinere Ergebnisse unserer Hirnforschung. J. Psychol. Neurol. 25, 279–468 (1919).

    Google Scholar 

  84. 84

    Nieuwenhuys, R., Broere, C.A. & Cerliani, L. A new myeloarchitectonic map of the human neocortex based on data from the Vogt-Vogt school. Brain Struct. Funct. 220, 2551–2573 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  85. 85

    Gordon, E.M. et al. Generation and evaluation of a cortical area parcellation from resting-state correlations. Cereb. Cortex 26, 288–303 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

  86. 86

    Power, J.D. et al. Functional network organization of the human brain. Neuron 72, 665–678 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  87. 87

    Yeo, B.T. et al. The organization of the human cerebral cortex estimated by intrinsic functional connectivity. J. Neurophysiol. 106, 1125–1165 (2011).

    Article  PubMed  PubMed Central  Google Scholar 

  88. 88

    Hacker, C.D. et al. Resting state network estimation in individual subjects. Neuroimage 82, 616–633 (2013). This study used a supervised classifier (multilayer perceptron) to identify resting-state networks in individual subjects. Classification was reliable in individuals not used for classifier training and generated more spatially specific resting-state networks than alternative methods tested.

    Article  PubMed  PubMed Central  Google Scholar 

  89. 89

    Cole, M.W., Bassett, D.S., Power, J.D., Braver, T.S. & Petersen, S.E. Intrinsic and task-evoked network architectures of the human brain. Neuron 83, 238–251 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  90. 90

    Fischl, B. et al. Automatically parcellating the human cerebral cortex. Cereb. Cortex 14, 11–22 (2004).

    Article  Google Scholar 

  91. 91

    Hodge, M.R. et al. ConnectomeDB—sharing human brain connectivity data. Neuroimage 124 Pt B, 1102–1107 (2016).

    Article  PubMed  Google Scholar 

  92. 92

    Marcus, D.S., Olsen, T.R., Ramaratnam, M. & Buckner, R.L. The Extensible Neuroimaging Archive Toolkit: an informatics platform for managing, exploring, and sharing neuroimaging data. Neuroinformatics 5, 11–34 (2007).

    Article  PubMed  Google Scholar 

  93. 93

    Smith, S.M. et al. A positive-negative mode of population covariation links brain connectivity, demographics and behavior. Nat. Neurosci. 18, 1565–1567 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  94. 94

    Yeo, B.T., Krienen, F.M., Chee, M.W. & Buckner, R.L. Estimates of segregation and overlap of functional connectivity networks in the human cerebral cortex. Neuroimage 88, 212–227 (2014).

    Article  PubMed  Google Scholar 

  95. 95

    Wang, D. et al. Parcellating cortical functional networks in individuals. Nat. Neurosci. 18, 1853–1860 (2015b).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  96. 96

    Hawrylycz, M. et al. Canonical genetic signatures of the adult human brain. Nat. Neurosci. 18, 1832–1844 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  97. 97

    Tavor, I. et al. Task-free MRI predicts individual differences in brain activity during task performance. Science 352, 216–220 (2016). The authors show that a model relating task-independent (rfMRI) measurements to task activity can accurately predict task activation maps for unseen subjects, suggesting a coupling between brain connectivity and function at the level of individual subjects.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  98. 98

    Reveley, C. et al. Superficial white matter fiber systems impede detection of long-range cortical connections in diffusion MR tractography. Proc. Natl. Acad. Sci. USA 112, E2820–E2828 (2015).

    Article  CAS  PubMed  Google Scholar 

  99. 99

    Dalcanton, J.J. 18 years of science with the Hubble Space Telescope. Nature 457, 41–50 (2009).

    Article  CAS  PubMed  Google Scholar 

  100. 100

    Kuhn, T. The Structure of Scientific Revolutions (University of Chicago Press, 1962).

  101. 101

    De Martino, F. et al. Frequency preference and attention effects across cortical depths in the human primary auditory cortex. Proc. Natl. Acad. Sci. USA 112, 16036–16041 (2015).

    Article  CAS  PubMed  Google Scholar 

  102. 102

    De Martino, F. et al. Spatial organization of frequency preference and selectivity in the human inferior colliculus. Nat. Commun. 4, 1386 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  103. 103

    Muckli, L. et al. Contextual feedback to superficial layers of V1. Curr. Biol. 25, 2690–2695 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  104. 104

    Malikovic, A. et al. Cytoarchitectonic analysis of the human extrastriate cortex in the region of V5/MT+: a probabilistic, stereotaxic map of area hOc5. Cereb. Cortex 17, 562–574 (2007).

    Article  PubMed  Google Scholar 

  105. 105

    Abdollahi, R.O. et al. Correspondences between retinotopic areas and myelin maps in human visual cortex. Neuroimage 99, 509–524 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  106. 106

    Murphy, K., Birn, R.M., Handwerker, D.A., Jones, T.B. & Bandettini, P.A. The impact of global signal regression on resting state correlations: are anti-correlated networks introduced? Neuroimage 44, 893–905 (2009).

    Article  PubMed  PubMed Central  Google Scholar 

  107. 107

    Power, J.D. et al. Methods to detect, characterize, and remove motion artifact in resting state fMRI. Neuroimage 84, 320–341 (2014).

    Article  PubMed  Google Scholar 

  108. 108

    Saad, Z.S. et al. Correcting brain-wide correlation differences in resting-state FMRI. Brain Connect. 3, 339–352 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  109. 109

    Gotts, S.J. et al. The perils of global signal regression for group comparisons: a case study of Autism Spectrum Disorders. Front. Hum. Neurosci. 7, 356 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  110. 110

    Saad, Z.S. et al. Trouble at rest: how correlation patterns and group differences become distorted after global signal regression. Brain Connect. 2, 25–32 (2012).

    Article  PubMed  PubMed Central  Google Scholar 

  111. 111

    Birn, R.M. The role of physiological noise in resting-state functional connectivity. Neuroimage 62, 864–870 (2012).

    Article  PubMed  PubMed Central  Google Scholar 

  112. 112

    Chang, C. & Glover, G.H. Relationship between respiration, end-tidal CO2, and BOLD signals in resting-state fMRI. Neuroimage 47, 1381–1393 (2009).

    Article  PubMed  PubMed Central  Google Scholar 

  113. 113

    Golestani, A.M., Chang, C., Kwinta, J.B., Khatamian, Y.B. & Jean Chen, J. Mapping the end-tidal CO2 response function in the resting-state BOLD fMRI signal: spatial specificity, test-retest reliability and effect of fMRI sampling rate. Neuroimage 104, 266–277 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  114. 114

    Smith, S.M. et al. Network modelling methods for FMRI. Neuroimage 54, 875–891 (2011).

    Article  PubMed  PubMed Central  Google Scholar 

  115. 115

    Larson-Prior, L.J. et al. Adding dynamics to the Human Connectome Project with MEG. Neuroimage 80, 190–201 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  116. 116

    Jack, C.R. Jr. et al. The Alzheimer's Disease Neuroimaging Initiative (ADNI): MRI methods. J. Magn. Reson. Imaging 27, 685–691 (2008).

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We thank the other investigators and staff members of the Human Connectome Project consortium for invaluable contributions to data acquisition, analysis and sharing. Additionally, we thank the many colleagues outside the HCP upon whose methodological contributions the paradigm espoused in this paper are also based. We thank S. Danker for assistance in manuscript preparation. Supported in part by the Human Connectome Project, WU-Minn-Ox Consortium (1U54MH091657) funded by the 16 NIH Institutes and Centers that support the NIH Blueprint for Neuroscience Research; the McDonnell Center for Systems Neuroscience at Washington University; and NIH F30 MH097312 (M.F.G.), RO1 MH-60974 (D.C.V.E.), P41 EB015894 (NIBIB; K.U.), Wellcome Trust 098369/Z/12/Z (S.M.S., J.L.R.A., T.E.J.B., M.J., E.C.R., S.N.S.), 5R01EB009352 (D.S.M.), 5P30NS048056 (D.S.M.) and 5R24MH108315 (D.S.M.).

Author information

Affiliations

Authors

Contributions

M.F.G., S.M.S., D.S.M., K.U. and D.C.V.E. framed the issues and generated the initial draft. M.F.G., S.M.S., D.S.M., J.L.R.A., E.J.A., T.E.J.B., T.S.C., M.P.H., M.J., S.M., E.C.R., S.N.S., J.X., E.Y., K.U. and D.C.V.E. contributed novel methods or analyses. M.F.G., S.M.S., D.S.M., T.E.J.B., T.S.C., M.P.H., E.C.R., S.N.S., J.X., E.Y., K.U. and D.C.V.E. wrote the manuscript.

Corresponding authors

Correspondence to Matthew F Glasser or David C Van Essen.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 Average cortical thickness map of 210 HCP subjects at each left hemisphere vertex and the associated colorized histogram.

The mean cortical thickness is around 2.6 mm, and this roughly divides fMRI data into high resolution (less than the mean cortical thickness) and low resolution (more than the mean cortical thickness). The HCP’s 3T and 7T chosen resolutions are also plotted. For the data in this figure and other Supplementary Figures, subject recruitment procedures and informed consent forms, including consent to share de-identified data, were approved by the Washington University institutional review board.

Supplementary Figure 2 dMRI data for 3T and 7T scans of the same subject (HCP Subject 158035).

Top row: Fractional anisotropy (FA) maps (axial, coronal and sagittal views). Notice that B1 inhomogeneities at 7T lead to poor SNR and noisy FA estimates in the inferior temporal regions (evident in the coronal views), but efforts have been taken to minimize them. Bottom row: DTI principal fiber orientations (coronal zoomed view of the area delineated by the yellow box). The orientations are RGB color-coded (Red: Left–Right, Green: Anterior–Posterior, Blue: Superior–Inferior) and superimposed on the structural T1w image. The pial surface and the WM/GM boundary surface are also shown. Reproduced, with permission, from Ref. 78.

Supplementary Figure 3 Patterns of cortico-striatal connectivity revealed by tractography.

Seed locations were in different cortical regions, including vmPFC (ventromedial prefrontal cortex), OFC (orbito-frontal cortex), dACC (dorsal anterior cingulate cortex), dPFC (dorsal prefrontal cortex) and Premotor cortex. Path probabilities (yellow: high, red: intermediate, black: low) are obtained using probabilistic tractography (FSL’s probtracx2, http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FDT/UserGuide) and the Matrix 3 (bidirectional white matter voxel to gray-matter terminations) algorithm to compute dense connectomes. N=150 HCP subjects were analyzed and averaged. Note the strong similarity to patterns of tracer-based connectivity reported in the macaque monkey, shown in the sketch on the right (reproduced with permission from Ref. 100).

Supplementary Figure 4 Multi-band imaging schematic and exemplar results.

Left: A schematic representation of the array coil elements over a human head image and a multi-band (MB) excitation (8 slices, in red). Each coil detects a linear combination of signals from each slice weighted by the sensitivity profile of that coil. Right. Four slices from a whole brain data set with standard acquisition (upper left, MB1) and with SMS/MB acquisition with MB=8, MB=12, and MB=16, all obtained using a 32-channel coil on a 3T scanner. Visual inspection of the M=16 image reveals discernible artifacts (e.g., ghosting), whereas the MB=12 and MB=8 images appear much cleaner. Quantitative estimates of cross slice contamination and reconstruction noise can be made. The HCP used MB=8 for fMRI and MB=3 for dMRI (now MB=4 or more for HCP Short).

Supplementary Figure 5 Beta map of the mean fMRI timeseries.

The timecourses were averaged over the whole brain (including gray matter, white matter, and CSF), regressed into the data of each subject, and then averaged across (n = 210 HCP subjects). One particularly striking characteristic of this map is how tissue-specific the global signal is (after ICA+FIX data cleanup), being generally most positive in grey matter, close to zero in white matter, and slightly negative in the ventricles. The tissue specificity of this signal argues against a non-physiologic, non-BOLD contrast-based cause of the signal (e.g., direct biophysical effects of subject head motion). This map by itself does not tell us to what extent the global signal is physiological noise vs. neural signal. Although the data are averaged in the volume across subjects, they still appear relatively sharp because they are not smoothed. Data at http://balsa.wustl.edu/0L1m.

Supplementary Figure 6 Visualization of the mean grey signal.

The mean grey timeseries beta map (top rows) and the ratio of the variance of the mean grey timeseries to the total BOLD variance (i.e., variance classified as signal by ICA+FIX; bottom rows). The absolute magnitude of the mean grey signal is highest in sensory regions including visual cortex, early somatosensory cortex (particularly of the face), early auditory cortex, and several thalamic nuclei including the LGN/MGN. Visual, somatosensory, auditory, and likely vestibular cortical areas are highlighted with black outlines. Data based on averaging across 210 HCP subjects, aligned using MSMAll for the cortical surface. When the global signal is strong in individual subjects, it closely matches the group average pattern, however when it is weaker, it may or may not match the group pattern. In these cases, the global signal often looks like one or another widespread RSN (e.g., the task positive or task negative (default mode) network). This is further evidence that removing the global signal as a preprocessing step may distort resting state functional connectivity and hence that we need a more selective way to clean global noise out of the data. The bottom row is a relative measure of how much the fMRI timeseries will be altered by regressing out the mean grey timeseries (on average across subjects), as it is a measure of the proportion of the total BOLD signal represented by the mean grey timeseries. Regressing out the mean grey signal will also tend to cause resting state gradient boundaries to move somewhat in the regions where this map has sharp gradients. Data at http://balsa.wustl.edu/2VnN.

Supplementary Figure 7 Effects of the Wishart rolloff on dense functional connectivity maps of both an individual subject and group data (210 HCP subjects; MSMAll surface registration).

Top rows show an individual subject, before (column 1) and after (column 2) Wishart rolloff for a seed location in lateral parietal cortex (white dot in upper left panels). The correlation increases dramatically as unstructured spatio-temporal noise is reduced, however the map is not substantially “smoothed” as it would be with typical smoothing algorithms. Bottom rows show a group dataset before and after Wishart rolloff for a seed location in the posterior cingulate sulcus (white dot in lower left panels). The dataset has been created using the MIGP algorithm to generate a group PCA series (d=4500) that represents the group concatenated timeseries. Because of the hard cutoff at PCA component number 4500, there is a ‘ringing” pattern resulting from spatial autocorrelation in the spatio-temporal noise that is represented by the PCA components with the lowest eigenvalues. If a Gaussian filter had been applied, this pattern of “local connectivity” would be a blob instead of rings. The Wishart rolloff eliminates these rings and again dramatically increases the SNR of the data. Data at http://balsa.wustl.edu/rrpl.

Supplementary Figure 8 The HCP language task (story vs baseline) beta maps and their spatial gradients.

Beta maps (rows 1 and 3) and gradient maps (rows 2 and 4) are from two independent groups of 210 HCP subjects, “210P” (rows 1 and 2) and “210V” (rows 3 and 4). Because of the large number of high quality HCP subjects, the beta maps are very similar across the two groups, and the strong gradients in the beta maps are also very similar. Also shown are white contours of a Bonferroni corrected significance threshold across all 91282 grayordinates (z+/- ~5). Two things are apparent: 1) Because of the large amount of high quality data, most of the brain is either significantly activated or deactivated (an issue that has been discussed before, see Ref. 60). Thus the statistical threshold is not particularly biologically meaningful (a point about statistical thresholds that generally applies). At the same time, the statistical threshold is also not strongly reproducible, in spite of the large amount of high quality data (highlighted ellipses show large differences in the area of activation classified as “significant” that are not particularly impressive when viewing the unthresholded beta maps). In contrast, the strong gradients in the beta maps are much more reproducible, are likely also more biologically meaningful, and hence provide a better substrate for defining regions of activation or comparing across studies. Data at http://balsa.wustl.edu/PrmK.

Supplementary Figure 9 Effects on average brain volume of registering 196 HCP brains to MNI space.

The total brain volume (minimal preprocessing pipelines’ whole brain mask) measured in the subject’s native physical space is around ~1350 cc; however, after registration to MNI space it is ~1800 cc, though the variability in brain volumes goes down as indicated by the narrower standard deviation error bars. This reduction in variability is the intended effect of registration, but the increase in brain volume is group average registration drift that was built into MNI standard space during the non-rigid registrations of the template generation process.

Supplementary Figure 10 A comparison between the HCP data and published retinotopic parcellation data.

Data from Ref. 64 and from 120 HCP subjects (from Q1-2) were registered using MSM areal-feature-based registration and group average registration drift was removed from both. Because of this, a contour in functional connectivity in the MT+ region distinguishing strong connectivity to the heavily myelinated IPS hotspot (LIPv, column 1) and to the STS (column 3) lines up with the border between MT+V4t (orange and yellow) and MST+FST (red and maroon, middle column). This illustrates the kind of precise cross-modal, cross-study boundary comparisons possible using the HCP-Style paradigm. Data at http://balsa.wustl.edu/x2Lz.

Supplementary Figure 11 Classification of area 55b in individual subjects by the areal classifier.

The typical location of area 55b is shown in black or white outlines on the inflated left hemisphere surface. The top two rows show a subject having an area 55b in the typical location found in the population. Rows 1 and 2 are entirely separate ‘test/retest’ runs of this subject through the full HCP MRI acquisition and analysis protocol. Column 1 is the subject’s individual curvature map, column 2 is the subject’s myelin map, column 3 is a d=40 RSN map that shows strong connectivity between 55b and other areas of the language network, column 4 is the raw probabilistic output of the areal classifier, column 5 is the final output of the areal classifier. Rows 3 and 4 show a different subject whose 55b is atypically split (heavy myelination running through the population average location of 55b and a concomitant lack of resting state connectivity). In both runs through the protocol, this subject shows a split 55b that is accurately detected by the areal classifier, showing that the classifier can accurately classify atypical subjects and that these atypical patterns are stable across time. Rows 5 and 6 show a third subject whose 55b is atypically shifted relative to its neighbors. Again the pattern is stable across time and the areal classifier is able to accurately detect the area. Reprinted from Ref. 34. Data at http://balsa.wustl.edu/WPPn.

Supplementary Figure 12 Effects of averaging surface coordinates and folding maps after areal feature-based registration (MSMAll).

The first row shows the group average midthickness surface (left), and the group average curvature map also displayed on inflated and flat surfaces (center and right). The group consisted of 210 subjects. The second row shows an individual subject’s midthickness surface (left), and the individual’s curvature map displayed also on inflated and flat surfaces (center and right). Note how much less detailed the group average surfaces and curvature maps are in most regions of cortex. However, in regions with consistent folding patterns across subjects and consistent relationships between cortical areas and folds, the group average patterns remain sharp (e.g. the central and calcarine sulci and the insula). The third row shows the group average T1w volume (after FNIRT nonlinear volume registration to MNI152 space of each subject in the group) together with the group average white (green) and pial (blue) surface contours. The fourth row shows the same individual subject’s T1w volume together with the individual’s white and pial surfaces (after aligning both the T1w volume and the surfaces to the group average in MNI space using FNIRT nonlinear volume registration). Despite the high precision of the white and pial surfaces in following the grey matter ribbon in the individual, the group average surfaces do not follow the group average volume particularly well, except in the regions where there are consistent folding patterns across subjects and consistent relationships between cortical areas and folds (as mentioned above). These issues also occur with folding-based surface registration (not shown), though they are less severe, because for folding-based registration the dominant factor is inconsistency in folding patterns across subjects (as no attempt is made to directly align cortical areas). The midthickness surfaces are the average of the white and pial surfaces (this average is performed on each individual, and the group midthickness surface is the average of the individual midthickness surfaces). Data at http://balsa.wustl.edu/7qP3.

Supplementary information

Supplementary Text and Figures

Supplementary Notes and Supplementary Figures 1–12 (PDF 3941 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Glasser, M., Smith, S., Marcus, D. et al. The Human Connectome Project's neuroimaging approach. Nat Neurosci 19, 1175–1187 (2016). https://doi.org/10.1038/nn.4361

Download citation

Further reading

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing