Brain Connectivity Studies on Structure-Function Relationships: A Short Survey with an Emphasis on Machine Learning

This short survey reviews the recent literature on the relationship between the brain structure and its functional dynamics. Imaging techniques such as diffusion tensor imaging (DTI) make it possible to reconstruct axonal fiber tracks and describe the structural connectivity (SC) between brain regions. By measuring fluctuations in neuronal activity, functional magnetic resonance imaging (fMRI) provides insights into the dynamics within this structural network. One key for a better understanding of brain mechanisms is to investigate how these fast dynamics emerge on a relatively stable structural backbone. So far, computational simulations and methods from graph theory have been mainly used for modeling this relationship. Machine learning techniques have already been established in neuroimaging for identifying functionally independent brain networks and classifying pathological brain states. This survey focuses on methods from machine learning, which contribute to our understanding of functional interactions between brain regions and their relation to the underlying anatomical substrate.

Visualization of antennal lobe glomeruli activated by nonappetitive D-limonene and appetitive 1-octen-3-ol odors via two types of olfactory organs in the blowfly Phormia regina

Appetite or feeding motivation relies significantly on food odors. In the blowfly Phormia regina, feeding motivation for sucrose is decreased by the odor of d-limonene but increased by the odor of 1-octen-3-ol odor. These flies have antennal lobes (ALs) consisting of several tens of glomerular pairs as a primary olfactory center in the brain. Odor information from different olfactory organs—specifically, the antennae and maxillary palps—goes to the corresponding glomeruli. To investigate how odors differently affect feeding motivation, we identified the olfactory organs and glomeruli that are activated by nonappetitive and appetitive odors. We first constructed a glomerular map of the antennal lobe in P. regina. Anterograde fluorescence labeling of antennal and maxillary afferent nerves, both of which project into the contralateral and ipsilateral ALs, revealed differential staining in glomerular regions. Some of the axonal fiber bundles from the antennae and maxillary palps projected to the subesophageal ganglion (SOG). We visualized the activation of the glomeruli in response to odor stimuli by immunostaining phosphorylated extracellular signal-regulated kinase (pERK). We observed different glomerulus activation under different odor stimulations. Referring to our glomerular map, we determined that antennal exposure to d-limonene odor activated the DA13 glomeruli, while exposure of the maxillary palps to 1-octen-3-ol activated the MxB1 glomeruli. Our results indicated that a nonappetitive odor input from the antennae and an appetitive odor input from the maxillary palps activate different glomeruli in the different regions of ALs in the blowfly P. regina. Collectively, our findings suggest that compartmentalization of glomeruli in AL is essential for proper transmission of odor information.

Corticospinal Wallerian Degeneration Before Myelination: A Case Report

Wallerian degeneration is defined as axonal fiber and myelin sheath degeneration that affects myelinated axons within the peripheral or central nervous system. Wallerian degeneration or anterograde axonal degeneration before myelination is rarely reported. Involvement of both corticospinal tracts (CSTs) is rarely documented in the literature. We present the postmortem neuropathologic findings of a 1-week-old male neonate born at 23 weeks of gestation with bilateral CST degeneration extending from the posterior limb of the internal capsule through the brainstem into the lumbar spinal cord. Abundant CD68- and CD163-positive macrophages were the prominent histopathology in both CSTs. The cerebrum, brainstem, and spinal cord were unmyelinated, as expected. In contrast, the spinal nerve roots demonstrated early myelination. This case illustrates that Wallerian degeneration occurs in unmyelinated axis cylinders.

Endogenous amylin contributes to birth of microglial cells in arcuate nucleus of hypothalamus and area postrema during fetal development

Amylin acts in the area postrema (AP) and arcuate nucleus (ARC) to control food intake. Amylin also increases axonal fiber outgrowth from the AP→nucleus tractus solitarius and from ARC→hypothalamic paraventricular nucleus. More recently, exogenous amylin infusion for 4 wk was shown to increase neurogenesis in adult rats in the AP. Furthermore, amylin has been shown to enhance leptin signaling in the ARC and ventromedial nucleus of the hypothalamus (VMN). Thus, we hypothesized that endogenous amylin could be a critical factor in regulating cell birth in the ARC and AP and that amylin could also be involved in the birth of leptin-sensitive neurons. Amylin+/− dams were injected with BrdU at embryonic day 12 and at postnatal day 2; BrdU+ cells were quantified in wild-type (WT) and amylin knockout (KO) mice. The number of BrdU+HuC/D+ neurons was similar in ARC and AP, but the number of BrdU+Iba1+ microglia was significantly decreased in both nuclei. Five-week-old WT and KO littermates were injected with leptin to test whether amylin is involved in the birth of leptin-sensitive neurons. Although there was no difference in the number of BrdU+c-Fos+ neurons in the ARC and dorsomedial nucleus, an increase in BrdU+c-Fos+ neurons was seen in VMN and lateral hypothalamus (LH) in amylin KO mice. In conclusion, these data suggest that during fetal development, endogenous amylin favors the birth of microglial cells in the ARC and AP and that it decreases the birth of leptin-sensitive neurons in the VMN and LH.

The Cerebral Cortex is Bisectionally Segregated into Two Fundamentally Different Functional Units of Gyri and Sulci

The human cerebral cortex is highly folded into diverse gyri and sulci. Accumulating evidences suggest that gyri and sulci exhibit anatomical, morphological, and connectional differences. Inspired by these evidences, we performed a series of experiments to explore the frequency-specific differences between gyral and sulcal neural activities from resting-state and task-based functional magnetic resonance imaging (fMRI) data. Specifically, we designed a convolutional neural network (CNN) based classifier, which can differentiate gyral and sulcal fMRI signals with reasonable accuracies. Further investigations of learned CNN models imply that sulcal fMRI signals are more diverse and more high frequency than gyral signals, suggesting that gyri and sulci truly play different functional roles. These differences are significantly associated with axonal fiber wiring and cortical thickness patterns, suggesting that these differences might be deeply rooted in their structural and cellular underpinnings. Further wavelet entropy analyses demonstrated the validity of CNN-based findings. In general, our collective observations support a new concept that the cerebral cortex is bisectionally segregated into 2 functionally different units of gyri and sulci.

Embedded finite elements for modeling axonal injury

The purpose of this paper is to propose and develop a large strain embedded finite element formulation that can be used to explicitly model axonal fiber bundle tractography from diffusion tensor imaging of the brain. Once incorporated, the fibers offer the capability to monitor tract-level strains that give insight into the biomechanics of brain injury. We show that one commercial software has a volume and mass redundancy issue when including embedded axonal fiber and that a newly developed algorithm is able to correct this discrepancy. We provide a validation analysis for stress and energy to demonstrate the method.

Computation of history-dependent mechanical damage of axonal fiber tracts in the brain: towards tracking sub-concussive and occupational damage to the brain

ABSTRACTFinite element models are frequently used to simulate traumatic brain injuries. However, current models are unable to capture the progressive damage caused by repeated head trauma. In this work, we propose a method for computing the history-dependent mechanical damage of axonal fiber bundle tracts in the brain. Through the introduction of multiple damage models, we provide the ability to link consecutive head impact simulations, so that potential injury to the brain can be tracked over time. In addition, internal damage variables are used to degrade the mechanical response of each axonal fiber bundle element. As a result, the stiffness of the aggregate tissue decreases as damage evolves. To counteract this degenerative process, we have also introduced a preliminary healing model that reverses the accumulated damage, based on a user-specified healing duration. Using two detailed examples, we demonstrate that damage produces a significant decrease in fiber stress, which ultimately propagates to the tissue level and produces a measurable decrease in overall stiffness. These results suggest that damage modeling has the potential to enhance current brain simulation techniques and lead to new insights, especially in the study of repetitive head injuries.