Periodic Reporting for period 1 - CoM-BraiN (Non-invasive Conduction Velocity Mapping in Brain Networks: A novel imaging framework for axonal fingerprinting of brain connections in health and disease)
Reporting period: 2022-11-01 to 2025-04-30
Brain function relies on communication between neuronal units at various levels, enabling our understanding, interactions, and decision-making in everyday life. This project focuses on understanding the anatomical microstructurealong the brain pathways that facilitate optimal signal transmission between functional regions of the brain.
Each pathway comprises thousands of axons, which act as communication cables, carrying signals from the neuron’s soma (cell body) to its dendrites. Axons vary in length, ranging from a few to tens of centimeters, depending on the distance between brain regions. Larger axon diameters allow faster signal transmission, but the primary signal booster—enhancing speed by up to five times—is the fatty myelin sheath that wraps around the axons. Within any given pathway, axons have a range of diameters, leading to varied transmission speeds based on their specific functional encoding.
Brain pathways are dynamic microstructures, adapting their signaling properties over a lifetime. Diseases such as Amyotrophic Lateral Sclerosis (ALS) and Multiple Sclerosis (MS) damage axons and myelin, altering their signaling abilities. Over time, this leads to functional and cognitive impairments as the diseases progress.
We use structural MRI (not functional MRI) to capture non-invasive 3D images that reveal changes in the microstructure of brain pathways. While structural MRI focuses on physical properties like axon diameters and myelin, we aim to predict functional signal transmission based on these structural scans. This prediction bridges the gap between structural changes and their potential impact on brain function.
By adapting structural MRI technology into a form of "in vivo microscopy" specifically sensitive to axon diameter and myelin changes, we aim to provide doctors with patient-specific insights for early detection of pathway dysfunction. Today, such anatomical information is typically only available via microscopic histology analysis of donated postmortem brains.
Each pathway comprises thousands of axons, which act as communication cables, carrying signals from the neuron’s soma (cell body) to its dendrites. Axons vary in length, ranging from a few to tens of centimeters, depending on the distance between brain regions. Larger axon diameters allow faster signal transmission, but the primary signal booster—enhancing speed by up to five times—is the fatty myelin sheath that wraps around the axons. Within any given pathway, axons have a range of diameters, leading to varied transmission speeds based on their specific functional encoding.
Brain pathways are dynamic microstructures, adapting their signaling properties over a lifetime. Diseases such as Amyotrophic Lateral Sclerosis (ALS) and Multiple Sclerosis (MS) damage axons and myelin, altering their signaling abilities. Over time, this leads to functional and cognitive impairments as the diseases progress.
We use structural MRI (not functional MRI) to capture non-invasive 3D images that reveal changes in the microstructure of brain pathways. While structural MRI focuses on physical properties like axon diameters and myelin, we aim to predict functional signal transmission based on these structural scans. This prediction bridges the gap between structural changes and their potential impact on brain function.
By adapting structural MRI technology into a form of "in vivo microscopy" specifically sensitive to axon diameter and myelin changes, we aim to provide doctors with patient-specific insights for early detection of pathway dysfunction. Today, such anatomical information is typically only available via microscopic histology analysis of donated postmortem brains.
Although structural MRI is limited to image resolution in the millimeter range and cannot directly measure individual axon diameters, its sensitivity to microstructural geometries goes beyond the image resolution. Indeed, MRI-based axon diameter methods have been presented over the last decade, but they still carry an inherent bias that limits their clinical application. We are working on a new type of axon diameter models that overcome these known biases to achieve a higher precision. However, we experienced that the classical approach to constructing biophysical models based on our anatmical assumptions becomes unreliable and unstable for this new type of axon model. The reason was related to the fact that our new axon model aimed at extracting more detailed anatomical information (i.e. more parameters) from the MRI signal. If the anatomical details had too similar MRI signal fingerprints, the model failed.
Instead, we applied machine learning (ML)-based models that are not limited by the assumptions in classical biophysical models, demonstrating outstanding performance—though their performance degrades dramatically when applied to real MRI data. Currently, we are testing a new type of data preprocessing step to overcome this degraded performance observed in the ML models, and the results are promising.
The next question is: how do we know what the next generation of axon models should look like when we cannot directly visualize axons and myelin? Specifically, this is an important question when training and testing ML-based models. We obtain this geometrical information using X-ray Nano Holotomography (XNH), a 3D imaging technique involving experimental setups such as beamlines installed at large-scale synchrotron facilities like the ID16A at ESRF in Grenoble, France, or the P10 at DESY in Hamburg, Germany. The imaging datasets are massive in size, and we have had to develop new image analysis strategies to handle and segment the collected XNH data into tissue classes containing axons, cells, blood vessels, and myelin, for subsequent quantification of the shape, size, and organization of these anatomical structures.
Using previously collected XNH image data of a monkey brain with 75-nanometer resolution, we have established such an image processing pipeline. This includes deep learning-based segmentation methods that, from a very sparse manually segmented 3D training dataset and without specific preprocessing of the data, can generalize the segmentation of entire XNH 3D image volumes into classes of axons, myelin, blood vessels, and cells. Now knowing how the real microstructure environment looks in 3D, we have developed a software tool named the White Matter Generator (WMG) (doi:10.3389/fninf.2024.1354708) to synthetically generate 3D meshes of realistic microstructure environments for diffusion MRI simulations. As illustrated in Figure 1, the WMG software can be used to generate various type of realistic 3D microstructural environments of both normal and diseased cases, such as demyelination or axonal degeneration, for training, testing, and validation of both ML and biophysical microstructure models.
The myelin wrapped around the axon is a key component that ensures the boosted conduction velocity of signal transmission along the axons. Today, MRI-based methods can provide insight into the amount of myelin within an MRI voxel. When combined with other MRI-based techniques, such as axon diameter methods, these can also provide insight into the so-called g-ratio—a single metric describing the ratio between axon diameter and myelin thickness. However, for these metrics it is challenging to differentiate whether a disease is degenerating the myelin or if the brain is in a re-establishment phase, repairing degenerated myelin after a disease attack—valuable information for doctors to better understand the state of such diseases.
We have explored a new myelin-based MRI method that potentially enables such insights into disease processes. Tests applied to animal models include induced focal demyelination, followed over time by remyelination. When compared with histological data from the same brain, these results show promising potential.
Instead, we applied machine learning (ML)-based models that are not limited by the assumptions in classical biophysical models, demonstrating outstanding performance—though their performance degrades dramatically when applied to real MRI data. Currently, we are testing a new type of data preprocessing step to overcome this degraded performance observed in the ML models, and the results are promising.
The next question is: how do we know what the next generation of axon models should look like when we cannot directly visualize axons and myelin? Specifically, this is an important question when training and testing ML-based models. We obtain this geometrical information using X-ray Nano Holotomography (XNH), a 3D imaging technique involving experimental setups such as beamlines installed at large-scale synchrotron facilities like the ID16A at ESRF in Grenoble, France, or the P10 at DESY in Hamburg, Germany. The imaging datasets are massive in size, and we have had to develop new image analysis strategies to handle and segment the collected XNH data into tissue classes containing axons, cells, blood vessels, and myelin, for subsequent quantification of the shape, size, and organization of these anatomical structures.
Using previously collected XNH image data of a monkey brain with 75-nanometer resolution, we have established such an image processing pipeline. This includes deep learning-based segmentation methods that, from a very sparse manually segmented 3D training dataset and without specific preprocessing of the data, can generalize the segmentation of entire XNH 3D image volumes into classes of axons, myelin, blood vessels, and cells. Now knowing how the real microstructure environment looks in 3D, we have developed a software tool named the White Matter Generator (WMG) (doi:10.3389/fninf.2024.1354708) to synthetically generate 3D meshes of realistic microstructure environments for diffusion MRI simulations. As illustrated in Figure 1, the WMG software can be used to generate various type of realistic 3D microstructural environments of both normal and diseased cases, such as demyelination or axonal degeneration, for training, testing, and validation of both ML and biophysical microstructure models.
The myelin wrapped around the axon is a key component that ensures the boosted conduction velocity of signal transmission along the axons. Today, MRI-based methods can provide insight into the amount of myelin within an MRI voxel. When combined with other MRI-based techniques, such as axon diameter methods, these can also provide insight into the so-called g-ratio—a single metric describing the ratio between axon diameter and myelin thickness. However, for these metrics it is challenging to differentiate whether a disease is degenerating the myelin or if the brain is in a re-establishment phase, repairing degenerated myelin after a disease attack—valuable information for doctors to better understand the state of such diseases.
We have explored a new myelin-based MRI method that potentially enables such insights into disease processes. Tests applied to animal models include induced focal demyelination, followed over time by remyelination. When compared with histological data from the same brain, these results show promising potential.
We have established the basic concepts of two methodological MRI-based novelties: a type of MRI-based axon diameter models and a novel MRI-based myelin metric. These will equip doctors and neuroscientists with a simple and precise tool to identify and understand how diseases impact pathways at their earliest stages, offering new perspectives on the mechanisms of brain function and disease progression. The natural key need is further research in the application to animal models mimicking different types of brain diseases for determining the specificity of the MRI methodologies. Then follows a translational stage to humans using clinical MRI scanners, with the expectation of internationalization for application in various brain diseases and for defining a standardization framework.
We have established a 3D deep learning-based segmentation framework for massive synchrotron XNH imaging datasets to be made publicly available. Through international collaborations, the accompanying segmentation results, as well as the training and test datasets, will be turned into a standardization framework, enabling a rich resource for training and testing future ML-based methods for semantic segmentation of 3D microstructure environments.
We have established a 3D deep learning-based segmentation framework for massive synchrotron XNH imaging datasets to be made publicly available. Through international collaborations, the accompanying segmentation results, as well as the training and test datasets, will be turned into a standardization framework, enabling a rich resource for training and testing future ML-based methods for semantic segmentation of 3D microstructure environments.