Saturday, Jul 22: 8:00 AM - 5:00 PM
2320
Educational Course - Full Day (8 hours)
Palais
Room: 516C
Brain white matter contains the highly myelinated axons that connect neurons in proximal or distant brain regions. These connections, organized in axon bundles, form the anatomic-functional networks that implement brain computations. Diffusion-weighted MRI (dMRI), together with computational tractography, provides in vivo estimates of the trajectory of these tracts through the white matter. In addition, dMRI provides information about the microscopic diffusion of water molecules within the tissue. Because water diffusion is affected by the physical structure of the tissue, this makes dMRI a sensitive probe of brain tissue properties. Taken together, these methods are used for tractometry, which focuses on the measurement of the physical properties of brain tracts. This course will introduce the concepts in tractometry, including basic data analysis and processing, segmentation of the white matter into anatomically-defined major tracts, the extraction of tissue properties along the tracts, and the statistical analysis of these properties. The course will also demonstrate the application of these methods to the understanding of the role of brain connections in brain development and aging, as well as in cognition. A broad range of approaches, methodologies and software tools will be presented, with significant hands-on components and interactive elements. This educational course is an updated and expanded version of the tractometry course which was very well-received and well-attended in 2021 and 2022. This iteration will include more introductory material, based on feedback from participants, and will spend more time discussing theory, methods, and applications, scaling up to a full-day course.
- Learners will understand the basic concepts of tract-based analysis of the human brain white matter.
- Learners will use software that segments white matter into different major bundles
- Learners will compare different approaches to statistical analysis of white matter tracts.
This course is intended for researchers (from trainees to faculty) with an interest in brain connectivity and the biology of brain connections. Researchers who are using datasets where multi-modal measurements are available (e.g., fMRI and dMRI) will benefit from expanding their analytic tool-set to include modern robust and rigorous tractometry methods.
Presentations
The last two decades of exploring the human brain's structure and function organization to unprecedented precision have opened up new challenges. Of these, the improvement of knowledge of brain anatomy is the most essential, especially considering the conceptual consequences this implies for the development of a reliable model of the human brain framework and its clinical and surgical applications. The neuroscience community has explicitly acknowledged the limitations of current tractography algorithms and, at the same time, highlighted the limits of our knowledge of the human brain white matter neuroanatomy. Historically, the latter was investigated by post-mortem cadaveric dissections, enabling a high-specific and systematic description of the different white matter pathways. This lecture will present the latest advances in diffusion tractography and dissection data fusion through the description of the main association pathways of the human brain.
The diffusion MRI signal is exquisitely sensitive to the underlying microstructure. By combining tailored MRI measurements and biophysical modeling of the diffusion process in white matter, specific microstructure features can be quantified in each voxel non-invasively. Accessible features include axonal water fraction, intra-axonal diffusivity, axonal orientation dispersion relative to the main bundle direction, and possibly axon diameters. This lecture will present relevant biophysical models of white matter and how to reliably estimate their parameters. Microstructure features can then be either quantified along an entire reconstructed bundle, or used to inform/constrain the reconstruction of realistic bundles in tractography.
Presenter
Ileana Jelescu, Laussane University Hospital Lausanne, AK
Switzerland
Tractography is the algorithm procedure that estimates axonal bundle trajectories from diffusion MRI data. A fundamental assumption of tractography is that water diffusion is faster along the axonal bundles than across them. Thus, the diffusion orientations of least-hindrance are approximations of the bundle orientations. Tractography methods use this assumption in different ways to follow bundle trajectories in the white matter and connect distant brain regions. This presentation will give an overview of tractography methods, and their use of diffusion MRI orientation estimates to reconstruct bundle trajectories. Moreover, we will discuss the current limitations of tractography methods and how to mitigate their effect using additional information, for instance, on the brain anatomy and morphology.
There are two broad families of methods for obtaining the white-matter bundles that undergo tractometry analyses. In the supervised approach, the goal is to reconstruct known tracts of interest (TOIs) based on a priori neuroanatomical definitions. In the unsupervised approach, the goal is to group whole-brain tractography data into clusters based on their similarity, without a priori knowledge of pathways of the brain. This presentation will give an overview of the two approaches, discuss how both of them can benefit by use of the underlying anatomy, and end with an interactive tour of white-matter tracts.
Tractography enables quantitative mapping of the brain’s structural connectivity using measures of connectivity or tissue microstructure. This presentation will provide a high-level overview of how tractography is used to perform quantitative analysis in health and disease. We will describe two main types of quantitative analyses of tractography, including: 1) tract-specific analysis that is typically hypothesis-driven and studies particular anatomical fiber tracts, and 2) connectome-based analysis that is often data-driven and generally studies the structural connectivity of the entire brain. We highlight three main processing steps that are common across most approaches for quantitative analysis of tractography, including methods for tractography correction, segmentation, and quantification. We then show example studies that have used these quantitative tractography approaches to study the brain’s white matter.
Tractography analysis often has separated steps of tractography mapping and tract analysis, but higher sensitivity and specificity could be achieved by integrating them together as novel tractography modalities. This hands-on section will demonstrate how differential tractography adopts a “tracking-the-differences” paradigm to track pathways with neuronal change in a patient. For group studies, I will demonstrate how correlational tractography can map fiber pathways correlated with a study variable by adopting a “tracking-the-correlation” paradigm. I will use a publicly available dataset of neurological disorder to show how differential and correlational tractography are used in practice.
Presenter
Fang-Cheng Yeh, Department of Neurological Surgery, University of Pittsburgh Medical Center Pittsburgh, USA
United States
Various dMRI models of white matter microstructure yield metrics relating to axon or “fiber” density. One such approach based on the reconstruction of fiber orientation distributions (FODs) from constrained spherical deconvolution (CSD) techniques, for example, allows obtaining a fiber-specific metric of “apparent fiber density”, which is typically analyzed using the fixel-based analysis (FBA) framework. In this presentation, I will first briefly describe the key concepts of the FBA framework. I will then focus on the challenges of interpretation of (fiber) density metrics. These challenges also extend to other dMRI models and analysis frameworks involving density metrics (e.g., neurite density from NODDI).
Presenter
Thijs Dhollander, Murdoch Children’s Research Institute Melbourne, Victoria
Australia
The human brain develops most rapidly during the first year of life, making early infancy a particularly exciting period for investigating structural properties of the white matter. Yet, those features that make infants unique, also bring distinct challenges for tractometry. For example, the almost complete lack of myelination in the infant brain results in reduced fractional anisotropy, and gray/white matter contrast compared to the adult brain. In this course, I will go over these challenges and propose specific adjustments that can be made to classical tractometry pipelines to improve the precision of white matter structural assessments in infants.
Presenter
Mareike Grotheer, Philipps-Universität Marburg
Psychology Department
Marburg
Germany
Postnatal brain development carries through infancy, adolescence, and into early adulthood. Development is varied and complex. Brain regions mature and reach developmental plateaus at differing rates, and microstructural features such as axonal density, diameter, fiber coherence, and myelination often follow different trajectories that are not consistent across tracts. Alongside this complex process, children are learning a vast array of cognitive and behavioral skills as they interact with their world. Together we will explore how the development of white matter relates to cognitive and behavioral outcomes, with a focus on reading, mental health, and the impact of adverse childhood experiences.
The brain undergoes significant alteration throughout the adult lifespan with significant deterioration of brain tissue in the later decades of life. However, there is great variability in neural senescence that is influenced by a range of factors. Demographics, genetics, health, and lifestyle all contribute to the complex effects of age on the brain. Diffusion imaging provides sensitive quantitative metrics of tissue microstructural properties and tractographic techniques to assess the anatomical properties of tissue deterioration. Here we review recent and classic studies using diffusion imaging in the assessment of factors that modulate the effects of aging on brain tissue, and the presumed functional consequences of these effects.
Presenter
David Salat, Massachusetts General Hospital Charlestown, MA
United States