Skip to main content

Imaging Brain Circuits to Decode Brain Computations: Multimodal Multiscale Imaging of Cortical Microcircuits to Model Predictive Coding in Human Vision

Periodic Reporting for period 4 - MULTICONNECT (Imaging Brain Circuits to Decode Brain Computations: Multimodal Multiscale Imaging of Cortical Microcircuits to Model Predictive Coding in Human Vision)

Reporting period: 2019-12-01 to 2020-11-30

The human brain is one of the largest and most complex biological networks known to exist. It consists of a large number of interacting circuits, which play a crucial role determining the computations it can perform and, thus, in enabling our cognitive abilities. The architecture of these circuits in humans, and therefore the computational basis of human cognition, remains largely unknown. The project focused on the multiscale imaging of human cortical microcircuits and on how these circuits contribute to computation and cognition. A structural and functional assessment of microcircuitry in the human brain only recently came within the realm of possibilities, thanks to the development of, first, optical tissue clearing and light sheet microscopy and, second, magnetic resonance imaging (MRI) at ultra-high field-strengths (UHF) of 7T and above. In-vivo functional imaging of human cortical activity, spatially resolved for cortical columns and layers, was also recently shown to be feasible with UHF fMRI. When UHF structural and functional MR imaging are combined, this has the exciting potential of imaging the connections and activity of different components within the cortical microcircuit.

The project concluded with the achievement of several imaging platforms, methodologies and data sets which achieved human cortical architecture and connectivity imaging and made large strides towards understanding human cortical computations. An ex vivo human brain MRI platform was developed with a suite of UHF imaging hardware and software for human brain tissue, which have led to new standards of quality and resolution of mesoscale ex vivo human brain tissue MRI scans for the investigation of cortical layer structure and white matter connections. A tissue clearing and light sheet fluorescence microscopy platform for human brain tissue was developed, which can image very large human brain tissue samples at microscopic resolution. These have created new standards of quality and field-of-view of ex vivo microscale human brain tissue microscopy for the investigation of cortical cytoarchitecture and microcircuitry. An in vivo MR imaging platform for white matter microstructure and high-resolution gray matter fMRI was created, including head coils tailored to high-resolution fMRI of the human visual cortex, methods for in vivo analysis of structural human brain connectivity and its functionally relevant microstructure (such as axonal diameters, density and myelination) with MRI, and an open-source analysis tool for analysis of human brain connectivity and microstructure with MRI.

Achieving the aims significantly advance our measurement of cortical architecture and understanding of how cortical microcircuits compute, provides important new reference data for graph analytical characterizations of the human connectome and generative models of human cortical dynamics, and informs modelling studies human cortical processing in health or after brain damage.
The mesoscale white matter connectivity between cortical visual system areas was investigated with newly developed ex vivo MRI devices and techniques. The goal of a human whole-brain mesoscale white matter connectome has been realized to great success by the development of a custom 9.4T whole human brain ex vivo RF-coil and the new kT-dSTEAM MRI pulse sequence (Fritz et al.). The goal of a human dorsal stream mesoscale white matter connectome was realized by the development of a custom 9.4T large sample ex vivo RF-coil, and the achievement of 60um isotropic human occipital lobe MRI (Sengupta et al.). These results were realized with essential contributions of in-house developed open-source analysis software (the Microstructure Diffusion Toolbox). The efforts, results and realization of bespoke 9.4T RF-coils and ground-breaking ex vivo MRI imaging results also lead to an authoritative review on the subject matter (Roebroeck et al.).

In working towards the goals of optimizing tissue clearing and deep fluorescence imaging and optically sectioned 3D fluorescence imaging, rapid technical developments in the field and local tests showed that availability of a novel light sheet microscope would be crucial to enable high resolution, high specificity, large field of view microscopy imaging. A new variation of lightsheet microscope, the ct-dSPIM was conceptualized and developed, which can image very large human brain tissue samples. An optical clearing (i.e. making transparent) and labelling protocol for cytoarchitecture characterization of human brain tissue samples was developed (MASH; Hildebrand et al. 2019). A further clearing and labeling approach in human cortical tissue, (hFRUIT; Hildebrand et al., 2020), and reported results highly relevant to the anatomical human microcircuit connectivity.

Several important methodological advances were made in the work on in vivo ultra-high field MRI goals. A set of specialized multi-transmit head coils tailored to sub-millimeter resolution fMRI of the human visual cortex were realized at 7T and 9.4T (Sengupta et al. 2016). Two methods for in vivo analysis of structural human brain connectivity and its functionally relevant microstructure (such as axonal diameters, density and myelination) with MRI in vivo were developed (De Santis et al.). Finally, the MESMERISED MRI imaging technique was developed for greatly accelerated ultra-high field MRI imaging for structural human brain connectivity and its functionally relevant microstructure (Fritz et al. 2020).
An ex vivo human brain MRI platform was developed with a suite custom 9.4T human brain tissue RF-coils and kT-dSTEAM, an new RF pulse sequence for post mortem human brain tissue diffusion MRI at 9.4T which achieved whole human brain anatomical MRI data at an unprecedented 75 micron isotropic resolution and diffusion MRI data at 400 micron isotropic resolution. These have led to new standards of quality and resolution of mesoscale ex vivo human brain tissue MRI scans for the investigation of cortical layer structure and white matter connections.

A tissue clearing and lightsheet fluorescence microscopy platform for human brain tissue was developed, including an optical tissue clearing and labelling protocol for cytoarchitecture characterization (MASH), a clearing and labeling approach to trace cortical microcircuit connections in human cortical tissue (hFRUIT), as well as a new variation of light sheet microscope, the ct-dSPIM, which can image very large human brain tissue samples. These have created new standards of quality and field-of-view of ex vivo microscale human brain tissue microscopy for the investigation of cortical cyto-architecture and microcircuitry.

An in vivo MR imaging platform for white matter microstructure and high-resolution gray matter fMRI was created, including head coils tailored to high-resolution fMRI of the human visual cortex, methods for in vivo analysis of structural human brain connectivity and its functionally relevant microstructure (such as axonal diameters, density and myelination) with MRI, and the MESMERISED MRI imaging technique which greatly accelerates ultra-high field MRI imaging for structural human brain connectivity and its functionally relevant microstructure. The Microstructure Diffusion Toolbox (MDT) was realized as an open-source analysis tool for analysis of human brain connectivity and microstructure with MRI (https://github.com/cbclab/MDT).
Multiscale imaging of human cortical architecture: from the entire occipital lobe to cortical layers