A machine learning approach for automated assessment of retinal vasculature in the oxygen induced retinopathy model

Animal Models of Retinopathy of Prematurity: Advances and Metabolic Regulators

Retinopathy of prematurity (ROP) is a primary cause of visual impairment and blindness in premature newborns, characterized by vascular abnormalities in the developing retina, with microvascular alteration, neovascularization, and in the most severe cases retinal detachment. To elucidate the pathophysiology and develop therapeutics for ROP, several pre-clinical experimental models of ROP were developed in different species. Among them, the oxygen-induced retinopathy (OIR) mouse model has gained the most popularity and critically contributed to our current understanding of pathological retinal angiogenesis and the discovery of potential anti-angiogenic therapies. A deeper comprehension of molecular regulators of OIR such as hypoxia-inducible growth factors including vascular endothelial growth factors as primary perpetrators and other new metabolic modulators such as lipids and amino acids influencing pathological retinal angiogenesis is also emerging, indicating possible targets for treatment strategies. This review delves into the historical progressions that gave rise to the modern OIR models with a focus on the mouse model. It also reviews the fundamental principles of OIR, recent advances in its automated assessment, and a selected summary of metabolic investigation enabled by OIR models including amino acid transport and metabolism.

Development of a Semi-automated Computer-based Tool for the Quantification of Vascular Tortuosity in the Murine Retina

Purpose

The murine oxygen-induced retinopathy (OIR) model is one of the most widely used animal models of ischemic retinopathy, mimicking hallmark pathophysiology of initial vaso-obliteration (VO) resulting in ischemia that drives neovascularization (NV). In addition to NV and VO, human ischemic retinopathies, including retinopathy of prematurity (ROP), are characterized by increased vascular tortuosity. Vascular tortuosity is an indicator of disease severity, need to treat, and treatment response in ROP. Current literature investigating novel therapeutics in the OIR model often report their effects on NV and VO, and measurements of vascular tortuosity are less commonly performed. No standardized quantification of vascular tortuosity exists to date despite this metric's relevance to human disease. This proof-of-concept study aimed to apply a previously published semi-automated computer-based image analysis approach (iROP-Assist) to develop a new tool to quantify vascular tortuosity in mouse models.

Design

Experimental study.

Subjects

C57BL/6J mice subjected to the OIR model.

Methods

In a pilot study, vasculature was manually segmented on flat-mount images of OIR and normoxic (NOX) mice retinas and segmentations were analyzed with iROP-Assist to quantify vascular tortuosity metrics. In a large cohort of age-matched (postnatal day 12 [P12], P17, P25) NOX and OIR mice retinas, NV, VO, and vascular tortuosity were quantified and compared. In a third experiment, vascular tortuosity in OIR mice retinas was quantified on P17 following intravitreal injection with anti-VEGF (aflibercept) or Immunoglobulin G isotype control on P12.

Main Outcome Measures

Vascular tortuosity.

Results

Cumulative tortuosity index was the best metric produced by iROP-Assist for discriminating between OIR mice and NOX controls. Increased vascular tortuosity correlated with disease activity in OIR. Treatment of OIR mice with aflibercept rescued vascular tortuosity.

Conclusions

Vascular tortuosity is a quantifiable feature of the OIR model that correlates with disease severity and may be quickly and accurately quantified using the iROP-Assist algorithm.

Financial Disclosures

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Applications of Deep Learning: Automated Assessment of Vascular Tortuosity in Mouse Models of Oxygen-Induced Retinopathy

Objective

To develop a generative adversarial network (GAN) to segment major blood vessels from retinal flat-mount images from oxygen-induced retinopathy (OIR) and demonstrate the utility of these GAN-generated vessel segmentations in quantifying vascular tortuosity.

Design

Development and validation of GAN.

Subjects

Three datasets containing 1084, 50, and 20 flat-mount mice retina images with various stains used and ages at sacrifice acquired from previously published manuscripts.

Methods

Four graders manually segmented major blood vessels from flat-mount images of retinas from OIR mice. Pix2Pix, a high-resolution GAN, was trained on 984 pairs of raw flat-mount images and manual vessel segmentations and then tested on 100 and 50 image pairs from a held-out and external test set, respectively. GAN-generated and manual vessel segmentations were then used as an input into a previously published algorithm (iROP-Assist) to generate a vascular cumulative tortuosity index (CTI) for 20 image pairs containing mouse eyes treated with aflibercept versus control.

Main Outcome Measures

Mean dice coefficients were used to compare segmentation accuracy between the GAN-generated and manually annotated segmentation maps. For the image pairs treated with aflibercept versus control, mean CTIs were also calculated for both GAN-generated and manual vessel maps. Statistical significance was evaluated using Wilcoxon signed-rank tests (P ≤ 0.05 threshold for significance).

Results

The dice coefficient for the GAN-generated versus manual vessel segmentations was 0.75 ± 0.27 and 0.77 ± 0.17 for the held-out test set and external test set, respectively. The mean CTI generated from the GAN-generated and manual vessel segmentations was 1.12 ± 0.07 versus 1.03 ± 0.02 (P = 0.003) and 1.06 ± 0.04 versus 1.01 ± 0.01 (P < 0.001), respectively, for eyes treated with aflibercept versus control, demonstrating that vascular tortuosity was rescued by aflibercept when quantified by GAN-generated and manual vessel segmentations.

Conclusions

GANs can be used to accurately generate vessel map segmentations from flat-mount images. These vessel maps may be used to evaluate novel metrics of vascular tortuosity in OIR, such as CTI, and have the potential to accelerate research in treatments for ischemic retinopathies.

Financial Disclosures

The author(s) have no proprietary or commercial interest in any materials discussed in this article.

Open-source analysis and visualization of segmented vasculature datasets with VesselVio

Vascular networks are fundamental components of biological systems. Quantitative analysis and observation of the features of these networks can improve our understanding of their roles in health and disease. Recent advancements in imaging technologies have enabled the generation of large-scale vasculature datasets, but barriers to analyzing these datasets remain. Modern analysis options are mainly limited to paid applications or open-source terminal-based software that requires programming knowledge with high learning curves. Here, we describe VesselVio, an open-source application developed to analyze and visualize pre-binarized vasculature datasets and pre-constructed vascular graphs. Vasculature datasets and graphs can be loaded with annotations and processed with custom parameters. Here, the program is tested on ground-truth datasets and is compared with current pipelines. The utility of VesselVio is demonstrated by the analysis of multiple formats of 2D and 3D datasets acquired with several imaging modalities, including annotated mouse whole-brain vasculature volumes.

Deep Learning-Based Detection of Endothelial Tip Cells in the Oxygen-Induced Retinopathy Model

Proliferative retinopathies, such as diabetic retinopathy and retinopathy of prematurity, are leading causes of vision impairment. A common feature is a loss of retinal capillary vessels resulting in hypoxia and neuronal damage. The oxygen-induced retinopathy model is widely used to study revascularization of an ischemic area in the mouse retina. The presence of endothelial tip cells indicates vascular recovery; however, their quantification relies on manual counting in microscopy images of retinal flat mount preparations. Recent advances in deep neural networks (DNNs) allow the automation of such tasks. We demonstrate a workflow for detection of tip cells in retinal images using the DNN-based Single Shot Detector (SSD). The SSD was designed for detection of objects in natural images. We adapt the SSD architecture and training procedure to the tip cell detection task and retrain the DNN using labeled tip cells in images of fluorescently stained retina flat mounts. Transferring knowledge from the pretrained DNN and extensive data augmentation reduced the amount of required labeled data. Our system shows a performance comparable to the human level, while providing highly consistent results. Therefore, such a system can automate counting of tip cells, a readout frequently used in retinopathy research, thereby reducing routine work for biomedical experts.

Big Data Research in Neuro-Ophthalmology: Promises and Pitfalls

BACKGROUND

Big data clinical research involves application of large data sets to the study of disease. It is of interest to neuro-ophthalmologists but also may be a challenge because of the relative rarity of many of the diseases treated.

EVIDENCE ACQUISITION

Evidence for this review was gathered from the authors' experiences performing analysis of large data sets and review of the literature.

RESULTS

Big data sets are heterogeneous, and include prospective surveys, medical administrative and claims data and registries compiled from medical records. High-quality studies must pay careful attention to aspects of data set selection, including potential bias, and data management issues, such as missing data, variable definition, and statistical modeling to generate appropriate conclusions. There are many studies of neuro-ophthalmic diseases that use big data approaches.

CONCLUSIONS

Big data clinical research studies complement other research methodologies to advance our understanding of human disease. A rigorous and careful approach to data set selection, data management, data analysis, and data interpretation characterizes high-quality studies.

Engineering the vasculature for islet transplantation

The microvasculature in the pancreatic islet is highly specialized for glucose sensing and insulin secretion. Although pancreatic islet transplantation is a potentially life-changing treatment for patients with insulin-dependent diabetes, a lack of blood perfusion reduces viability and function of newly transplanted tissues. Functional vasculature around an implant is not only necessary for the supply of oxygen and nutrients but also required for rapid insulin release kinetics and removal of metabolic waste. Inadequate vascularization is particularly a challenge in islet encapsulation. Selectively permeable membranes increase the barrier to diffusion and often elicit a foreign body reaction including a fibrotic capsule that is not well vascularized. Therefore, approaches that aid in the rapid formation of a mature and robust vasculature in close proximity to the transplanted cells are crucial for successful islet transplantation or other cellular therapies. In this paper, we review various strategies to engineer vasculature for islet transplantation. We consider properties of materials (both synthetic and naturally derived), prevascularization, local release of proangiogenic factors, and co-transplantation of vascular cells that have all been harnessed to increase vasculature. We then discuss the various other challenges in engineering mature, long-term functional and clinically viable vasculature as well as some emerging technologies developed to address them. The benefits of physiological glucose control for patients and the healthcare system demand vigorous pursuit of solutions to cell transplant challenges. STATEMENT OF SIGNIFICANCE: Insulin-dependent diabetes affects more than 1.25 million people in the United States alone. Pancreatic islets secrete insulin and other endocrine hormones that control glucose to normal levels. During preparation for transplantation, the specialized islet blood vessel supply is lost. Furthermore, in the case of cell encapsulation, cells are protected within a device, further limiting delivery of nutrients and absorption of hormones. To overcome these issues, this review considers methods to rapidly vascularize sites and implants through material properties, pre-vascularization, delivery of growth factors, or co-transplantation of vessel supporting cells. Other challenges and emerging technologies are also discussed. Proper vascular growth is a significant component of successful islet transplantation, a treatment that can provide life-changing benefits to patients.

Diffuse and nonlinear imaging of multiscale vascular parameters for in vivo monitoring of preclinical mammary tumors

Diffuse optical imaging (DOI) techniques provide a wide-field or macro assessment of the functional tumor state and have shown substantial promise for monitoring treatment efficacy in cancer. Conversely, intravital microscopy provides a high-resolution view of the tumor state and has played a key role in characterizing treatment response in the preclinical setting. There has been little prior work in investigating how the macro and micro spatial scales can be combined to develop a more comprehensive and translational view of treatment response. To address this, a new multiscale preclinical imaging technique called diffuse and nonlinear imaging (DNI) was developed. DNI combines multiphoton microscopy with spatial frequency domain imaging (SFDI) to provide multiscale data sets of tumor microvascular architecture coregistered within wide-field hemodynamic maps. A novel method was developed to match the imaging depths of both modalities by utilizing informed SFDI spatial frequency selection. An in vivo DNI study of murine mammary tumors revealed multiscale relationships between tumor oxygen saturation and microvessel diameter, and tumor oxygen saturation and microvessel length (|Pearson's ρ| ≥ 0.5, P < 0.05). Going forward, DNI will be uniquely enabling for the investigation of multiscale relationships in tumors during treatment.

Automatic analysis of the retinal avascular area in the rat oxygen-induced retinopathy model

Purpose

The aim of this study was to create an algorithm to automate, accelerate, and standardize the process of avascular area segmentation in images from a rat oxygen-induced retinopathy (OIR) model.

Methods

Within 6 h of birth, full-term pups born to Sprague Dawley rat dams that had undergone partial bilateral uterine artery ligation at embryonic day 19.5 were placed into a controlled oxygen environment (Oxycycler, BioSpherix, Parish, NY) at 50% oxygen for 48 h, followed by cycling between 10% and 50% oxygen every 24 h until day 15. The pups were then moved into room air until day 18.5. Ten lectin-stained retinal flat mounts were imaged in montage fashion at 10x magnification. Three masked human reviewers measured two parameters, total retinal area and peripheral avascular area, for each image using the ImageJ freehand selection tool. The outputs of each read were measured as number of pixels. The gold standard value for each image was the mean of the three human reads. Interrater agreement for the measurement of total retinal area, avascular area, and percent avascular area was calculated using type A intraclass correlation coefficients (ICCs) with a two-way random effects model. Automated avascular area identification (A3ID) is a method written in ImageJ Macro that is intended for use in the Fiji (Fiji is Just ImageJ) image processing platform. The input for A3ID is a rat retinal image, and the output is the avascular area (in pixels). A3ID utilizes a random forest classifier with a connected-components algorithm and post-processing filters for size and shape. A separate algorithm calculates the total retinal area. We compared the output of both algorithms to gold standard measurements by calculating ICCs, performing linear regression, and determining the Dice coefficients for both algorithms. We also constructed a Bland-Altman plot for A3ID output.

Results

The ICC for percent peripheral avascular/total area between human readers was 0.995 (CI: 0.974-0.999), with p<0.001. The ICC between A3ID and the gold standard was calculated for three image parameters-avascular area: 0.974 (CI: 0.899-0.993), with p<0.001; total retinal area: 0.465 (CI: 0.0-0.851), with p=0.001; and the percent peripheral avascular/total area: 0.94 (CI: 0.326-0.989), with p<0.001. In the linear regression analysis, the slope for prediction of the gold standard percent peripheral avascular/total area from A3ID was 0.98, with R2=0.975. A3ID and the total retinal area algorithm achieve an average Dice coefficient of 0.891 and 0.952, respectively. The Bland-Altman analysis revealed a trend for computer underestimation of the peripheral avascular area in images with low peripheral avascular area and overestimation of peripheral avascular area in images with large peripheral avascular areas.

Conclusions

A3ID reliably predicts peripheral avascular area based on rat OIR retinal images. When the peripheral avascular area is particularly high or low, hand segmentation of images may be superior.

Vertebral artery fusiform aneurysm geometry in predicting rupture risk

Cerebral aneurysms affect a significant portion of the adult population worldwide. Despite significant progress, the development of robust techniques to evaluate the risk of aneurysm rupture remains a critical challenge. We hypothesize that vertebral artery fusiform aneurysm (VAFA) morphology may be predictive of rupture risk and can serve as a deciding factor in clinical management. To investigate the VAFA morphology, we use a combination of image analysis and machine learning techniques to study a geometric feature set computed from a depository of 37 (12 ruptured and 25 un-ruptured) aneurysm images. Of the 571 unique features we compute, we distinguish five features for use by our machine learning classification algorithm by an analysis of statistical significance. These machine learning methods achieve state-of-the-art classification performance (81.43 ± 13.08%) for the VAFA morphology, and identify five features (cross-sectional area change of aneurysm, maximum diameter of nearby distal vessel, solidity of aneurysm, maximum curvature of nearby distal vessel, and ratio of curvature between aneurysm and its nearby proximal vessel) as effective predictors of VAFA rupture risk. These results suggest that the geometric features of VAFA morphology may serve as useful non-invasive indicators for the prediction of aneurysm rupture risk in surgical settings.