Segmentation of anatomical layers and imaging artifacts in intravascular polarization sensitive optical coherence tomography using attending physician and boundary cardinality losses

Intravascular ultrasound and optical coherence tomography are widely available for assessing coronary stenoses and provide critical information to optimize percutaneous coronary intervention. Intravascular polarization-sensitive optical coherence tomography (PS-OCT) measures the polarization state of the light scattered by the vessel wall in addition to conventional cross-sectional images of subsurface microstructure. This affords reconstruction of tissue polarization properties and reveals improved contrast between the layers of the vessel wall along with insight into collagen and smooth muscle content. Here, we propose a convolutional neural network model, optimized using two new loss terms (Boundary Cardinality and Attending Physician), that takes advantage of the additional polarization contrast and classifies the lumen, intima, and media layers in addition to guidewire and plaque shadows. Our model segments the media boundaries through fibrotic plaques and continues to estimate the outer media boundary behind shadows of lipid-rich plaques. We demonstrate that our multi-class classification model outperforms existing methods that exclusively use conventional OCT data, predominantly segment the lumen, and consider subsurface layers at most in regions of minimal disease. Segmentation of all anatomical layers throughout diseased vessels may facilitate stent sizing and will enable automated characterization of plaque polarization properties for investigation of the natural history and significance of coronary atheromas.

Vascular oxidative stress causes neutrophil arrest in brain capillaries, leading to decreased cerebral blood flow and contributing to memory impairment in a mouse model of Alzheimer’s disease

INTRODUCTION

In this study, we explore the role of oxidative stress produced by NOX2-containing NADPH oxidase as a molecular mechanism causing capillary stalling and cerebral blood flow deficits in the APP/PS1 mouse model of AD.

METHODS

We inhibited NOX2 in APP/PS1 mice by administering a 10 mg/kg dose of the peptide inhibitor gp91-ds-tat i.p., for two weeks. We used in vivo two-photon imaging to measure capillary stalling, penetrating arteriole flow, and vascular inflammation. We also characterized short-term memory function and gene expression changes in cerebral microvessels.

RESULTS

We found that after NOX2 inhibition capillary stalling, as well as parenchymal and vascular inflammation, were significantly reduced. In addition, we found a significant increase in penetrating arteriole flow, followed by an improvement in short-term memory, and downregulation of inflammatory gene expression pathways.

DISCUSSION

Oxidative stress is a major mechanism leading to microvascular dysfunction in AD, and represents an important therapeutic target.

VEGF signalling causes stalls in brain capillaries and reduces cerebral blood flow in Alzheimer's mice

Increased incidence of stalled capillary blood flow caused by adhesion of leucocytes to the brain microvascular endothelium leads to a 17% reduction of cerebral blood flow and exacerbates short-term memory loss in multiple mouse models of Alzheimer’s disease.

Here, we report that vascular endothelial growth factor (VEGF) signalling at the luminal side of the brain microvasculature plays an integral role in the capillary stalling phenomenon of the APP/PS1 mouse model.

Administration of the anti-mouse VEGF-A164 antibody, an isoform that inhibits blood–brain barrier hyperpermeability, reduced the number of stalled capillaries within an hour of injection, leading to an immediate increase in average capillary blood flow but not capillary diameter. VEGF-A inhibition also reduced the overall endothelial nitric oxide synthase protein concentrations, increased occludin levels and decreased the penetration of circulating Evans Blue dye across the blood–brain barrier into the brain parenchyma, suggesting increased blood–brain barrier integrity. Capillaries prone to neutrophil adhesion after anti-VEGF-A treatment also had lower occludin concentrations than flowing capillaries.

Taken together, our findings demonstrate that VEGF-A signalling in APP/PS1 mice contributes to aberrant endothelial nitric oxide synthase /occludin-associated blood–brain barrier permeability, increases the incidence of capillary stalls, and leads to reductions in cerebral blood flow. Reducing leucocyte adhesion by inhibiting luminal VEGF signalling may provide a novel and well-tolerated strategy for improving brain microvascular blood flow in Alzheimer’s disease patients.

A pilot study investigating the effects of voluntary exercise on capillary stalling and cerebral blood flow in the APP/PS1 mouse model of Alzheimer's disease

Exercise exerts a beneficial effect on the major pathological and clinical symptoms associated with Alzheimer’s disease in humans and mouse models of the disease. While numerous mechanisms for such benefits from exercise have been proposed, a clear understanding of the causal links remains elusive. Recent studies also suggest that cerebral blood flow in the brain of both Alzheimer’s patients and mouse models of the disease is decreased and that the cognitive symptoms can be improved when blood flow is restored. We therefore hypothesized that the mitigating effect of exercise on the development and progression of Alzheimer’s disease may be mediated through an increase in the otherwise reduced brain blood flow. To test this idea, we performed a pilot study to examine the impact of three months of voluntary wheel running in a small cohort of ~1-year-old APP/PS1 mice on short-term memory function, brain inflammation, amyloid deposition, and baseline cerebral blood flow. Our findings that exercise led to a trend toward improved spatial short-term memory, reduced brain inflammation, markedly increased neurogenesis in the dentate gyrus, and a reduction in hippocampal amyloid-beta deposits are consistent with other reports on the impact of exercise on the progression of Alzheimer’s related symptoms in mouse models. Notably, we did not observe any impact of wheel running on overall baseline blood flow nor on the incidence of non-flowing capillaries, a mechanism we recently identified as one contributing factor to cerebral blood flow deficits in mouse models of Alzheimer’s disease. Overall, our findings add to the emerging picture of differential effects of exercise on cognition and blood flow in Alzheimer’s disease pathology by showing that capillary stalling is not decreased following exercise.

Increasing cerebral blood flow improves cognition into late stages in Alzheimer's disease mice

Alzheimer's disease is associated with a 20-30% reduction in cerebral blood flow. In the APP/PS1 mouse model of Alzheimer's disease, inhibiting neutrophil adhesion using an antibody against the neutrophil specific protein Ly6G was recently shown to drive rapid improvements in cerebral blood flow that was accompanied by an improvement in performance on short-term memory tasks. Here, in a longitudinal aging study, we assessed how far into disease development a single injection of anti-Ly6G treatment can acutely improve short-term memory function. We found that APP/PS1 mice as old as 15-16 months had improved performance on the object replacement and Y-maze tests of spatial and working short-term memory, measured at one day after anti-Ly6G treatment. APP/PS1 mice at 17-18 months of age or older did not show acute improvements in cognitive performance, although we did find that capillary stalls were still reduced and cerebral blood flow was still increased by 17% in 21-22-months-old APP/PS1 mice given anti-Ly6G antibody. These data add to the growing body of evidence suggesting that cerebral blood flow reductions are an important contributing factor to the cognitive dysfunction associated with neurodegenerative disease. Thus, interfering with neutrophil adhesion could be a new therapeutic approach for Alzheimer's disease.

High fat diet worsens Alzheimer's disease-related behavioral abnormalities and neuropathology in APP/PS1 mice, but not by synergistically decreasing cerebral blood flow

Obesity is linked to increased risk for and severity of Alzheimer's disease (AD). Cerebral blood flow (CBF) reductions are an early feature of AD and are also linked to obesity. We recently showed that non-flowing capillaries, caused by adhered neutrophils, contribute to CBF reduction in mouse models of AD. Because obesity could exacerbate the vascular inflammation likely underlying this neutrophil adhesion, we tested links between obesity and AD by feeding APP/PS1 mice a high fat diet (Hfd) and evaluating behavioral, physiological, and pathological changes. We found trends toward poorer memory performance in APP/PS1 mice fed a Hfd, impaired social interactions with either APP/PS1 genotype or a Hfd, and synergistic impairment of sensory-motor function in APP/PS1 mice fed a Hfd. The Hfd led to increases in amyloid-beta monomers and plaques in APP/PS1 mice, as well as increased brain inflammation. These results agree with previous reports showing obesity exacerbates AD-related pathology and symptoms in mice. We used a crowd-sourced, citizen science approach to analyze imaging data to determine the impact of the APP/PS1 genotype and a Hfd on capillary stalling and CBF. Surprisingly, we did not see an increase in the number of non-flowing capillaries or a worsening of the CBF deficit in APP/PS1 mice fed a Hfd as compared to controls, suggesting that capillary stalling is not a mechanistic link between a Hfd and increased severity of AD in mice. Reducing capillary stalling by blocking neutrophil adhesion improved CBF and short-term memory function in APP/PS1 mice, even when fed a Hfd.

A topological encoding convolutional neural network for segmentation of 3D multiphoton images of brain vasculature using persistent homology

The clinical evidence suggests that cognitive disorders are associated with vasculature dysfunction and decreased blood flow in the brain. Hence, a functional understanding of the linkage between brain functionality and the vascular network is essential. However, methods to systematically and quantitatively describe and compare structures as complex as brain blood vessels are lacking. 3D imaging modalities such as multiphoton microscopy enables researchers to capture the network of brain vasculature with high spatial resolutions. Nonetheless, image processing and inference are some of the bottlenecks for biomedical research involving imaging, and any advancement in this area impacts many research groups. Here, we propose a topological encoding convolutional neural network based on persistent homology to segment 3D multiphoton images of brain vasculature. We demonstrate that our model out-performs state-of-the-art models in terms of the Dice coefficient and it is comparable in terms of other metrics such as sensitivity. Additionally, the topological characteristics of our model's segmentation results mimic manual ground truth. Our code and model are open source at https://github.com/mhaft/DeepVess.

Brain Capillary Networks Across Species: A few Simple Organizational Requirements Are Sufficient to Reproduce Both Structure and Function

Despite the key role of the capillaries in neurovascular function, a thorough characterization of cerebral capillary network properties is currently lacking. Here, we define a range of metrics (geometrical, topological, flow, mass transfer, and robustness) for quantification of structural differences between brain areas, organs, species, or patient populations and, in parallel, digitally generate synthetic networks that replicate the key organizational features of anatomical networks (isotropy, connectedness, space-filling nature, convexity of tissue domains, characteristic size). To reach these objectives, we first construct a database of the defined metrics for healthy capillary networks obtained from imaging of mouse and human brains. Results show that anatomical networks are topologically equivalent between the two species and that geometrical metrics only differ in scaling. Based on these results, we then devise a method which employs constrained Voronoi diagrams to generate 3D model synthetic cerebral capillary networks that are locally randomized but homogeneous at the network-scale. With appropriate choice of scaling, these networks have equivalent properties to the anatomical data, demonstrated by comparison of the defined metrics. The ability to synthetically replicate cerebral capillary networks opens a broad range of applications, ranging from systematic computational studies of structure-function relationships in healthy capillary networks to detailed analysis of pathological structural degeneration, or even to the development of templates for fabrication of 3D biomimetic vascular networks embedded in tissue-engineered constructs.

Deep convolutional neural networks for segmenting 3D in vivo multiphoton images of vasculature in Alzheimer disease mouse models

The health and function of tissue rely on its vasculature network to provide reliable blood perfusion. Volumetric imaging approaches, such as multiphoton microscopy, are able to generate detailed 3D images of blood vessels that could contribute to our understanding of the role of vascular structure in normal physiology and in disease mechanisms. The segmentation of vessels, a core image analysis problem, is a bottleneck that has prevented the systematic comparison of 3D vascular architecture across experimental populations. We explored the use of convolutional neural networks to segment 3D vessels within volumetric in vivo images acquired by multiphoton microscopy. We evaluated different network architectures and machine learning techniques in the context of this segmentation problem. We show that our optimized convolutional neural network architecture with a customized loss function, which we call DeepVess, yielded a segmentation accuracy that was better than state-of-the-art methods, while also being orders of magnitude faster than the manual annotation. To explore the effects of aging and Alzheimer's disease on capillaries, we applied DeepVess to 3D images of cortical blood vessels in young and old mouse models of Alzheimer's disease and wild type littermates. We found little difference in the distribution of capillary diameter or tortuosity between these groups, but did note a decrease in the number of longer capillary segments (>75μm) in aged animals as compared to young, in both wild type and Alzheimer's disease mouse models.