Automatic segmentation, feature extraction and comparison of healthy and stroke cerebral vasculature

Reconstructing blood flow in data-poor regimes: a vasculature network kernel for Gaussian process regression

Blood flow reconstruction in the vasculature is important for many clinical applications. However, in clinical settings, the available data are often quite limited. For instance, transcranial Doppler ultrasound is a non-invasive clinical tool that is commonly used in clinical settings to measure blood velocity waveforms at several locations. This amount of data is grossly insufficient for training machine learning surrogate models, such as deep neural networks or Gaussian process regression. In this work, we propose a Gaussian process regression approach based on empirical kernels constructed by data generated from physics-based simulations-enabling near-real-time reconstruction of blood flow in data-poor regimes. We introduce a novel methodology to reconstruct the kernel within the vascular network. The proposed kernel encodes both spatiotemporal and vessel-to-vessel correlations, thus enabling blood flow reconstruction in vessels that lack direct measurements. We demonstrate that any prediction made with the proposed kernel satisfies the conservation of mass principle. The kernel is constructed by running stochastic one-dimensional blood flow simulations, where the stochasticity captures the epistemic uncertainties, such as lack of knowledge about boundary conditions and uncertainties in vasculature geometries. We demonstrate the performance of the model on three test cases, namely, a simple Y-shaped bifurcation, abdominal aorta and the circle of Willis in the brain.

Deep Learning for 3D Vascular Segmentation in Phase Contrast Tomography

Automated blood vessel segmentation is critical for biomedical image analysis, as vessel morphology changes are associated with numerous pathologies. Still, precise segmentation is difficult due to the complexity of vascular structures, anatomical variations across patients, the scarcity of annotated public datasets, and the quality of images. Our goal is to provide a foundation on the topic and identify a robust baseline model for application to vascular segmentation using a new imaging modality, Hierarchical Phase-Contrast Tomography (HiP-CT). We begin with an extensive review of current machine learning approaches for vascular segmentation across various organs. Our work introduces a meticulously curated training dataset, verified by double annotators, consisting of vascular data from three kidneys imaged using Hierarchical Phase-Contrast Tomography (HiP-CT) as part of the Human Organ Atlas Project. HiP-CT, pioneered at the European Synchrotron Radiation Facility in 2020, revolutionizes 3D organ imaging by offering resolution around 20μm/voxel, and enabling highly detailed localized zooms up to 1μm/voxel without physical sectioning. We leverage the nnU-Net framework to evaluate model performance on this high-resolution dataset, using both known and novel samples, and implementing metrics tailored for vascular structures. Our comprehensive review and empirical analysis on HiP-CT data sets a new standard for evaluating machine learning models in high-resolution organ imaging. Our three experiments yielded Dice scores of 0.9523 and 0.9410, and 0.8585, respectively. Nevertheless, DSC primarily assesses voxel-to-voxel concordance, overlooking several crucial characteristics of the vessels and should not be the sole metric for deciding the performance of vascular segmentation. Our results show that while segmentations yielded reasonably high scores-such as centerline Dice values ranging from 0.82 to 0.88, certain errors persisted. Specifically, large vessels that collapsed due to the lack of hydro-static pressure (HiP-CT is an ex vivo technique) were segmented poorly. Moreover, decreased connectivity in finer vessels and higher segmentation errors at vessel boundaries were observed. Such errors, particularly in significant vessels, obstruct the understanding of the structures by interrupting vascular tree connectivity. Through our review and outputs, we aim to set a benchmark for subsequent model evaluations using various modalities, especially with the HiP-CT imaging database.

Static and dynamic analysis of cerebral blood flow in fifty-six large arterial vessel networks

Objective.The cerebral vasculature is formed of an intricate network of blood vessels over many different length scales. Changes in their structure and connection are implicated in multiple cerebrovascular and neurological disorders. In this study, we present a novel approach to the quantitative analysis of the cerebral macrovasculature using computational and mathematical tools in a large dataset.Approach.We analysed a publicly available vessel dataset from a cohort of 56 (32/24F/M) healthy subjects. This dataset includes digital reconstructions of human brain macrovasculatures. We then propose a new mathematical model to compute blood flow dynamics and pressure distributions within these 56-representative cerebral macrovasculatures and quantify the results across this cohort.Main results.Statistical analysis showed that the steady state level of cerebrovascular resistance (CVR) gradually increases with age in both men and women. These age-related changes in CVR are in good agreement with previously reported values. All subjects were found to have only small phase angles (<6°) between blood pressure and blood flow at the cardiac frequency.Significance.These results showed that the dynamic component of blood flow adds very little phase shift at the cardiac frequency, which implies that the cerebral macrocirculation can be regarded as close to steady state in its behaviour, at least in healthy populations, irrespective of age or sex. This implies that the phase shift observed in measurements of blood flow in cerebral vessels is caused by behaviour further down the vascular bed. This behaviour is important for future statistical models of the dynamic maintenance of oxygen and nutrient supply to the brain.

Accurate and robust segmentation of cerebral vasculature on four-dimensional arterial spin labeling magnetic resonance angiography using machine-learning approach

Segmentation of cerebral vasculature on MR vascular images is of great significance for clinical application and research. However, the existing cerebrovascular segmentation approaches are limited due to insufficient image contrast and complicated algorithms. This study aims to explore the potential of the emerging four-dimensional arterial spin labeling magnetic resonance angiography (4D ASL-MRA) technique for fast and accurate cerebrovascular segmentation with a simple machine-learning approach. Nine temporal features were extracted from the intensity-time signal of each voxel, and eight spatial features from the neighboring voxels. Then, the unsupervised outlier detection algorithm, i.e. Isolation Forest, is used for segmentation of the vascular voxels based on the extracted features. The total length of the centerlines of the intracranial arterial vasculature, the dice similarity coefficient (DSC), and the average Hausdorff Distance (AVGHD) on the cross-sections of small- to large-sized vessels were calculated to evaluate the performance of the segmentation approach on 4D ASL-MRA of 18 subjects. Experiments show that the temporal information on 4D ASL-MRA can largely improve the segmentation performance. In addition, the proposed segmentation approach outperforms the traditional methods that were performed on the 3D image (i.e. the temporal average intensity projection of 4D ASL-MRA) and the previously proposed frame-wise approach. In conclusion, this study demonstrates that accurate and robust segmentation of cerebral vasculature is achievable on 4D ASL-MRA by using a simple machine-learning approach with appropriate features.

Associations of Intracranial Artery Length and Branch Number on Time-of-Flight MRA With Cognitive Impairment in Hypertensive Older Males

BACKGROUND

Hypertension-induced impairment of the cerebral artery network contributes to cognitive impairment. Characterizing the structure and function of cerebral arteries may facilitate the understanding of hypertension-related pathological mechanisms and lead to the development of new indicators for cognitive impairment.

PURPOSE

To investigate the associations between morphological features of the intracranial arteries distal to the circle of Willis on time-of-flight MRA (TOF-MRA) and cognitive performance in a hypertensive cohort.

STUDY TYPE

Prospective observational study.

POPULATION

189 hypertensive older males (mean age 64.9 ± 7.2 years).

FIELD STRENGTH/SEQUENCE

TOF-MRA sequence with a 3D spoiled gradient echo readout and arterial spin labeling perfusion imaging sequence with a 3D stack-of-spirals fast spin echo readout at 3T.

ASSESSMENT

The intracranial arteries were segmented from TOF-MRA and the total length of distal arteries (TLoDA) and number of arterial branches (NoB) were calculated. The mean gray matter cerebral blood flow (GM-CBF) was extracted from arterial spin labeling perfusion imaging. The cognitive level was assessed with short-term and long-term delay-recall auditory verbal learning test (AVLT) scores, and with montreal cognitive assessment.

STATISTICAL TESTS

Univariable and multivariable linear regression were used to analyze the associations between TLoDA, NoB, GM-CBF and the cognitive assessment scores, with P < 0.05 indicating significance.

RESULTS

TLoDA (r = 0.314) and NoB (r = 0.346) were significantly correlated with GM-CBF. Multivariable linear regression analyses showed that TLoDA and NoB, but not GM-CBF (P = 0.272 and 0.141), were significantly associated with short-term and long-term delay-recall AVLT scores. These associations remained significant after adjusting for GM-CBF.

DATA CONCLUSION

The TLoDA and NoB of distal intracranial arteries on TOF-MRA are significantly associated with cognitive impairment in hypertensive subjects.

LEVEL OF EVIDENCE

2 TECHNICAL EFFICACY: Stage 3.

End to end stroke triage using cerebrovascular morphology and machine learning

Background

Rapid and accurate triage of acute ischemic stroke (AIS) is essential for early revascularization and improved patient outcomes. Response to acute reperfusion therapies varies significantly based on patient-specific cerebrovascular anatomy that governs cerebral blood flow. We present an end-to-end machine learning approach for automatic stroke triage.

Methods

Employing a validated convolutional neural network (CNN) segmentation model for image processing, we extract each patient's cerebrovasculature and its morphological features from baseline non-invasive angiography scans. These features are used to detect occlusion's presence and the site automatically, and for the first time, to estimate collateral circulation without manual intervention. We then use the extracted cerebrovascular features along with commonly used clinical and imaging parameters to predict the 90 days functional outcome for each patient.

Results

The CNN model achieved a segmentation accuracy of 94% based on the Dice similarity coefficient (DSC). The automatic stroke detection algorithm had a sensitivity and specificity of 92% and 94%, respectively. The models for occlusion site detection and automatic collateral grading reached 96% and 87.2% accuracy, respectively. Incorporating the automatically extracted cerebrovascular features significantly improved the 90 days outcome prediction accuracy from 0.63 to 0.83.

Conclusion

The fast, automatic, and comprehensive model presented here can improve stroke diagnosis, aid collateral assessment, and enhance prognostication for treatment decisions, using cerebrovascular morphology.

Ultrasound, photoacoustic, and magnetic resonance imaging to study hyperacute pathophysiology of traumatic and vascular brain injury

BACKGROUND AND PURPOSE

Cerebrovascular dynamics and pathomechanisms that evolve in the minutes and hours following traumatic vascular injury in the brain remain largely unknown. We investigated the pathophysiology evolution in mice within the first 3 hours after closed-head traumatic brain injury (TBI) and subarachnoid hemorrhage (SAH), two significant traumatic vascular injuries.

METHODS

We took a multimodal imaging approach using photoacoustic imaging, color Doppler ultrasound, and MRI to track injury outcomes using a variety of metrics.

RESULTS

Brain oxygenation and velocity-weighted volume of blood flow (VVF) values significantly decreased from baseline to 15 minutes after both TBI and SAH. TBI resulted in 19.2% and 41.0% ipsilateral oxygenation and VVF reductions 15 minutes postinjury, while SAH resulted in 43.9% and 85.0% ipsilateral oxygenation and VVF reduction (p < .001). We found partial recovery of oxygenation from 15 minutes to 3 hours after injury for TBI but not SAH. Hemorrhage, edema, reduced perfusion, and altered diffusivity were evident from MRI scans acquired 90-150 minutes after injury in both injury models, although the spatial distribution was mostly focal for TBI and diffuse for SAH.

CONCLUSIONS

The results reveal that the cerebral oxygenation deficits immediately following injuries are reversible for TBI and irreversible for SAH. Our findings can inform future studies on mitigating these early responses to improve long-term recovery.

A network-based model of dynamic cerebral autoregulation

Cerebrovascular diseases continue to be one of the leading causes of morbidity and mortality in humans. Abnormalities in dynamic cerebral autoregulation (dCA) have been implicated in many of these disease conditions. Accurate models are therefore needed to better understand the complex pathophysiology behind impaired dCA. We thus present here a simple framework for modelling a vessel-driven network model of dCA in the microvasculature, as opposed to the conventional compartmental modelling approach. Network models incorporate the actual connectivity and anatomy of the vasculature, thereby allowing us to include and trace changes in the calibre and morphology of individual vessels, investigate the spatial specificity and heterogeneity of the various control mechanisms to help disentangle their contributions, and link the model parameters to the actual network physiology. The proposed control feedback mechanisms are incorporated at the level of the individual vessel, and the dynamic pressure and flow fields are solved for here within a simple vessel network. In response to an upstream pressure drop, the network is found to be able to recover cerebral blood flow (CBF) while exhibiting the characteristic autoregulatory behaviour in terms of changes in vessel calibre and the biphasic flow response. We assess the feasibility of our formulation in larger networks by comparing the simulation results to those obtained using a one-dimensional (1D) model of CBF applied to the same microvasculature network and find that our model results are in very good agreement with the 1D solution, while significantly reducing the computational cost, thus enabling more detailed models of network behaviour to be adopted in the future. Accurate and computationally feasible models of dCA that are more representative of the vasculature can help increase the translatability of haemodynamic models into the clinical environment, which would help develop more informed treatment guidelines for patients with cerebrovascular diseases.

A dataset of rodent cerebrovasculature from in vivo multiphoton fluorescence microscopy imaging

We present MiniVess, the first annotated dataset of rodent cerebrovasculature, acquired using two-photon fluorescence microscopy. MiniVess consists of 70 3D image volumes with segmented ground truths. Segmentations were created using traditional image processing operations, a U-Net, and manual proofreading. Code for image preprocessing steps and the U-Net are provided. Supervised machine learning methods have been widely used for automated image processing of biomedical images. While much emphasis has been placed on the development of new network architectures and loss functions, there has been an increased emphasis on the need for publicly available annotated, or segmented, datasets. Annotated datasets are necessary during model training and validation. In particular, datasets that are collected from different labs are necessary to test the generalizability of models. We hope this dataset will be helpful in testing the reliability of machine learning tools for analyzing biomedical images.

Vascular perfusion differs in two distinct PDGFRβ-positive zones within the ischemic core of male mice 2 weeks following photothrombotic stroke

Stroke therapy has largely focused on preventing damage and encouraging repair outside the ischemic core, as the core is considered irreparable. Recently, several studies have suggested endogenous responses within the core are important for limiting the spread of damage and enhancing recovery, but the role of blood flow and capillary pericytes in this process is unknown. Using the Rose Bengal photothrombotic model of stroke, we illustrate blood vessels are present in the ischemic core and peri-lesional regions 2 weeks post stroke in male mice. A FITC-albumin gel cast of the vasculature revealed perfusion of these vessels, suggesting cerebral blood flow (CBF) may be partially present, without vascular leakage. The length of these vessels is significantly reduced compared to uninjured regions, but the average width is greater, suggesting they are either larger vessels that survived the initial injury, smaller vessels that have expanded in size (i.e., arteriogenesis), or that neovascularization begins with larger vessels. Concurrently, we observed an increase in platelet-derived growth factor receptor beta (PDGFRβ, a marker of pericytes) expression within the ischemic core in two distinct patterns, one which resembles pericyte-derived fibrotic scarring at the edge of the core, and one which is vessel associated and may represent blood vessel recovery. We find little evidence for dividing cells on these intralesional blood vessels 2 weeks post stroke. Our study provides evidence flow is present in PDGFRβ-positive vessels in the ischemic core 2 weeks post stroke. We hypothesize intralesional CBF is important for limiting injury and for encouraging endogenous repair following cerebral ischemia.

Physics-Informed Neural Networks for Brain Hemodynamic Predictions Using Medical Imaging

Determining brain hemodynamics plays a critical role in the diagnosis and treatment of various cerebrovascular diseases. In this work, we put forth a physics-informed deep learning framework that augments sparse clinical measurements with one-dimensional (1D) reduced-order model (ROM) simulations to generate physically consistent brain hemodynamic parameters with high spatiotemporal resolution. Transcranial Doppler (TCD) ultrasound is one of the most common techniques in the current clinical workflow that enables noninvasive and instantaneous evaluation of blood flow velocity within the cerebral arteries. However, it is spatially limited to only a handful of locations across the cerebrovasculature due to the constrained accessibility through the skull's acoustic windows. Our deep learning framework uses in vivo real-time TCD velocity measurements at several locations in the brain combined with baseline vessel cross-sectional areas acquired from 3D angiography images and provides high-resolution maps of velocity, area, and pressure in the entire brain vasculature. We validate the predictions of our model against in vivo velocity measurements obtained via four-dimensional (4D) flow magnetic resonance imaging (MRI) scans. We then showcase the clinical significance of this technique in diagnosing cerebral vasospasm (CVS) by successfully predicting the changes in vasospastic local vessel diameters based on corresponding sparse velocity measurements. We show this capability by generating synthetic blood flow data after cerebral vasospasm at various levels of stenosis. Here, we demonstrate that the physics-based deep learning approach can estimate and quantify the subject-specific cerebral hemodynamic variables with high accuracy despite lacking knowledge of inlet and outlet boundary conditions, which is a significant limitation for the accuracy of the conventional purely physics-based computational models.

Novel imaging markers for altered cerebrovascular morphology in aging, stroke, and Alzheimer's disease

BACKGROUND AND PURPOSE

Altered brain vasculature is a key phenomenon in several neurologic disorders. This paper presents a quantitative assessment of the anatomical variations in the Circle of Willis (CoW) and vascular morphology in healthy aging, acute ischemic stroke (AIS) and Alzheimer's Disease (AD).

METHODS

We used our novel automatic method to segment and extract geometric features of the cerebral vasculature from MR angiography scans of 175 healthy subjects, which were used to create a probabilistic atlas of cerebrovasculature and to study normal aging and intersubject variations in CoW anatomy. Subsequently, we quantified and analyzed vascular alterations in 45AIS and 50 AD patients, two prominent cerebrovascular and neurodegenerative disorders.

RESULTS

In the sampled cohort, we determined that the CoW is fully formed in only 35% of healthy adults and found significantly (p < .05) increased tortuosity and fractality, with increasing age and also with disease in both AIS and AD. We also found significantly lower vessel length, volume, and number of branches in AIS patients, as expected. The AD cerebral vessels exhibited significantly smaller diameter and more complex branching patterns, compared to age-matched healthy adults. These changes were significantly heightened (p < .05) among healthy, early onset mild AD, and moderate/severe dementia groups.

CONCLUSION

Although our study does not include longitudinal data due to paucity of such datasets, the specific geometric features and quantitative comparisons demonstrate the potential for using vascular morphology as a noninvasive imaging biomarker for neurologic disorders.

Simultaneous vessel segmentation and unenhanced prediction using self-supervised dual-task learning in 3D CTA (SVSUP)

BACKGROUND AND OBJECTIVE

The vessel segmentation in CT angiography (CTA) provides an important basis for automatic diagnosis and hemodynamics analysis. Virtual unenhanced (VU) CT images obtained by dual-energy CT can assist clinical diagnosis and reduce radiation dose by obviating true unenhanced imaging (UECT). However, accurate segmentation of all vessels in the head-neck CTA (HNCTA) remains a challenge, and VU images are currently not available from conventional single-energy CT imaging.

METHODS

In this paper, we proposed a self-supervised dual-task deep learning strategy to fully automatically segment all vessels and predict unenhanced CT images from single-energy HNCTA based on a developed iterative residual-sharing scheme. The underlying idea was to use the correlation between the two tasks to improve task performance while avoiding manual annotation for model training.

RESULTS

The feasibility of the strategy was verified using the data of 24 patients. For vessel segmentation task, the proposed model achieves a significantly higher average Dice coefficient (84.83%, P-values 10^-3 in paired t-test) than the state-of-the-art segmentation model, vanilla VNet (78.94%), and several popular 3D vessel segmentation models, including Hessian-matrix based filter (62.59%), optically-oriented flux (66.33%), spherical flux model (66.91%), and deep vessel net (66.47%). For the unenhanced prediction task, the average ROI-based error compared to the UECT in the artery tissue is 6.1±4.5 HU, similar to previously reported 6.4±5.1 HU for VU reconstruction.

CONCLUSIONS

Results show that the proposed dual-task framework can effectively improve the accuracy of vessel segmentation in HNCTA, and it is feasible to predict the unenhanced image from single-energy CTA, providing a potential new approach for radiation dose saving. Moreover, to our best knowledge, this is the first reported annotation-free deep learning-based full-image vessel segmentation for HNCTA.

Mapping physiological and pathological functions of cortical vasculature through aggregation-induced emission nanoprobes assisted quantitative, in vivo NIR-II imaging

Cerebrovascular disease includes all disorders that affect cerebrovascular and cerebral circulation. Unfortunately, there is currently a lack of a systematic method to image blood vessels directly and achieve accurate quantification. Herein, we build a non-invasive, quantitative imaging and characterization system applicable to mapping physiological and pathological functions of cortical vasculature. Assisted by aggregation-induced emission (AIE) luminogens with either excitation or emission at near-infrared-II (NIR-II) region, large-depth and/or high signal-to-background ratio images of cerebral blood vessels from mice and marmosets are captured, based on which we develop an optical metric of vessel thickness in an automated, pixel-wise manner and both two-dimensional (2D) and three-dimensional (3D) contexts. By monitoring time-dependent cerebrovascular images in marmosets, periodic changes in the diameter of vibrating cerebral blood vessels are found to be regulated mainly by heartbeat. In mice photothrombosis model, vessel alterations throughout the whole process of thrombotic stroke are found to be stage-dependent. From a large field of view, the distance-dependent vessel thickness variation before and right after stroke is obtained away from the thrombus site. Importantly, a buffer zone exists right surrounding the lesion, indicating the inhomogeneity of vascular morphological changes. Biologically excretable AIE nanoparticles are used for assessing physiological and pathological functions, offering great potential for clinical translation.

An evaluation of performance measures for arterial brain vessel segmentation

BACKGROUND

Arterial brain vessel segmentation allows utilising clinically relevant information contained within the cerebral vascular tree. Currently, however, no standardised performance measure is available to evaluate the quality of cerebral vessel segmentations. Thus, we developed a performance measure selection framework based on manual visual scoring of simulated segmentation variations to find the most suitable measure for cerebral vessel segmentation.

METHODS

To simulate segmentation variations, we manually created non-overlapping segmentation errors common in magnetic resonance angiography cerebral vessel segmentation. In 10 patients, we generated a set of approximately 300 simulated segmentation variations for each ground truth image. Each segmentation was visually scored based on a predefined scoring system and segmentations were ranked based on 22 performance measures common in the literature. The correlation of visual scores with performance measure rankings was calculated using the Spearman correlation coefficient.

RESULTS

The distance-based performance measures balanced average Hausdorff distance (rank = 1) and average Hausdorff distance (rank = 2) provided the segmentation rankings with the highest average correlation with manual rankings. They were followed by overlap-based measures such as Dice coefficient (rank = 7), a standard performance measure in medical image segmentation.

CONCLUSIONS

Average Hausdorff distance-based measures should be used as a standard performance measure in evaluating cerebral vessel segmentation quality. They can identify more relevant segmentation errors, especially in high-quality segmentations. Our findings have the potential to accelerate the validation and development of novel vessel segmentation approaches.