Skeletonization method for vessel delineation of arteriovenous malformation

Application of artificial intelligence in brain arteriovenous malformations: Angioarchitectures, clinical symptoms and prognosis prediction

BACKGROUND

Artificial intelligence (AI) has rapidly advanced in the medical field, leveraging its intelligence and automation for the management of various diseases. Brain arteriovenous malformations (AVM) are particularly noteworthy, experiencing rapid development in recent years and yielding remarkable results. This paper aims to summarize the applications of AI in the management of AVMs management.

METHODS

Literatures published in PubMed during 1999-2022, discussing AI application in AVMs management were reviewed.

RESULTS

AI algorithms have been applied in various aspects of AVM management, particularly in machine learning and deep learning models. Automatic lesion segmentation or delineation is a promising application that can be further developed and verified. Prognosis prediction using machine learning algorithms with radiomic-based analysis is another meaningful application.

CONCLUSIONS

AI has been widely used in AVMs management. This article summarizes the current research progress, limitations and future research directions.

Reconstructing microvascular network skeletons from 3D images: What is the ground truth?

Structural changes to microvascular networks are increasingly highlighted as markers of pathogenesis in a wide range of disease, e.g. Alzheimer's disease, vascular dementia and tumour growth. This has motivated the development of dedicated 3D imaging techniques, alongside the creation of computational modelling frameworks capable of using 3D reconstructed networks to simulate functional behaviours such as blood flow or transport processes. Extraction of 3D networks from imaging data broadly consists of two image processing steps: segmentation followed by skeletonisation. Much research effort has been devoted to segmentation field, and there are standard and widely-applied methodologies for creating and assessing gold standards or ground truths produced by manual annotation or automated algorithms. The Skeletonisation field, however, lacks widely applied, simple to compute metrics for the validation or optimisation of the numerous algorithms that exist to extract skeletons from binary images. This is particularly problematic as 3D imaging datasets increase in size and visual inspection becomes an insufficient validation approach. In this work, we first demonstrate the extent of the problem by applying 4 widely-used skeletonisation algorithms to 3 different imaging datasets. In doing so we show significant variability between reconstructed skeletons of the same segmented imaging dataset. Moreover, we show that such a structural variability propagates to simulated metrics such as blood flow. To mitigate this variability we introduce a new, fast and easy to compute super metric that compares the volume, connectivity, medialness, bifurcation point identification and homology of the reconstructed skeletons to the original segmented data. We then show that such a metric can be used to select the best performing skeletonisation algorithm for a given dataset, as well as to optimise its parameters. Finally, we demonstrate that the super metric can also be used to quickly identify how a particular skeletonisation algorithm could be improved, becoming a powerful tool in understanding the complex implication of small structural changes in a network.

Comparative Study of Automated Algorithms for Brain Arteriovenous Malformation Nidus Extent Identification Using 3DRA

PURPOSE

When performing a brain arteriovenous malformation (bAVMs) intervention, computer-assisted analysis of bAVMs can aid clinicians in planning precise therapeutic alternatives. Therefore, we aim to assess currently available methods for bAVMs nidus extent identification over 3DRA. To this end, we establish a unified framework to contrast them over the same dataset, fully automatising the workflows.

MATERIALS AND METHODS

We retrospectively collected contrast-enhanced 3DRA scans of patients with bAVMs. A segmentation network was used to automatically acquire the brain vessels segmentation for each case. We applied the nidus extent identification algorithms over each of the segmentations, computing overlap measurements against manual nidus delineations.

RESULTS

We evaluated the methods over a private dataset with 22 3DRA scans of individuals with bAVMs. The best-performing alternatives resulted in [Formula: see text] and [Formula: see text] dice coefficient values.

CONCLUSIONS

The mathematical morphology-based approach showed higher robustness through inter-case variability. The skeleton-based approach leverages the skeleton topomorphology characteristics, while being highly sensitive to anatomical variations and the skeletonisation method employed. Overall, nidus extent identification algorithms are also limited by the quality of the raw volume, as the consequent imprecise vessel segmentation will hinder their results. Performance of the available alternatives remains subpar. This analysis allows for a better understanding of the current limitations.

Automatic Detection and Segmentation of Thrombi in Abdominal Aortic Aneurysms Using a Mask Region-Based Convolutional Neural Network with Optimized Loss Functions

The detection and segmentation of thrombi are essential for monitoring the disease progression of abdominal aortic aneurysms (AAAs) and for patient care and management. As they have inherent capabilities to learn complex features, deep convolutional neural networks (CNNs) have been recently introduced to improve thrombus detection and segmentation. However, investigations into the use of CNN methods is in the early stages and most of the existing methods are heavily concerned with the segmentation of thrombi, which only works after they have been detected. In this work, we propose a fully automated method for the whole process of the detection and segmentation of thrombi, which is based on a well-established mask region-based convolutional neural network (Mask R-CNN) framework that we improve with optimized loss functions. The combined use of complete intersection over union (CIoU) and smooth L1 loss was designed for accurate thrombus detection and then thrombus segmentation was improved with a modified focal loss. We evaluated our method against 60 clinically approved patient studies (i.e., computed tomography angiography (CTA) image volume data) by conducting 4-fold cross-validation. The results of comparisons to multiple other state-of-the-art methods suggested the superior performance of our method, which achieved the highest F1 score for thrombus detection (0.9197) and outperformed most metrics for thrombus segmentation.

Segmentation techniques of brain arteriovenous malformations for 3D visualization: a systematic review

BACKGROUND

Visualization, analysis and characterization of the angioarchitecture of a brain arteriovenous malformation (bAVM) present crucial steps for understanding and management of these complex lesions. Three-dimensional (3D) segmentation and 3D visualization of bAVMs play hereby a significant role. We performed a systematic review regarding currently available 3D segmentation and visualization techniques for bAVMs.

METHODS

PubMed, Embase and Google Scholar were searched to identify studies reporting 3D segmentation techniques applied to bAVM characterization. Category of input scan, segmentation (automatic, semiautomatic, manual), time needed for segmentation and 3D visualization techniques were noted.

RESULTS

Thirty-three studies were included. Thirteen (39%) used MRI as baseline imaging modality, 9 used DSA (27%), and 7 used CT (21%). Segmentation through automatic algorithms was used in 20 (61%), semiautomatic segmentation in 6 (18%), and manual segmentation in 7 (21%) studies. Median automatic segmentation time was 10 min (IQR 33), semiautomatic 25 min (IQR 73). Manual segmentation time was reported in only one study, with the mean of 5-10 min. Thirty-two (97%) studies used screens to visualize the 3D segmentations outcomes and 1 (3%) study utilized a heads-up display (HUD). Integration with mixed reality was used in 4 studies (12%).

CONCLUSIONS

A golden standard for 3D visualization of bAVMs does not exist. This review describes a tendency over time to base segmentation on algorithms trained with machine learning. Unsupervised fuzzy-based algorithms thereby stand out as potential preferred strategy. Continued efforts will be necessary to improve algorithms, integrate complete hemodynamic assessment and find innovative tools for tridimensional visualization.

Scalable robust graph and feature extraction for arbitrary vessel networks in large volumetric datasets

BACKGROUND

Recent advances in 3D imaging technologies provide novel insights to researchers and reveal finer and more detail of examined specimen, especially in the biomedical domain, but also impose huge challenges regarding scalability for automated analysis algorithms due to rapidly increasing dataset sizes. In particular, existing research towards automated vessel network analysis does not always consider memory requirements of proposed algorithms and often generates a large number of spurious branches for structures consisting of many voxels. Additionally, very often these algorithms have further restrictions such as the limitation to tree topologies or relying on the properties of specific image modalities.

RESULTS

We propose a scalable iterative pipeline (in terms of computational cost, required main memory and robustness) that extracts an annotated abstract graph representation from the foreground segmentation of vessel networks of arbitrary topology and vessel shape. The novel iterative refinement process is controlled by a single, dimensionless, a-priori determinable parameter.

CONCLUSIONS

We are able to, for the first time, analyze the topology of volumes of roughly 1 TB on commodity hardware, using the proposed pipeline. We demonstrate improved robustness in terms of surface noise, vessel shape deviation and anisotropic resolution compared to the state of the art. An implementation of the presented pipeline is publicly available in version 5.1 of the volume rendering and processing engine Voreen.

Recent progress understanding pathophysiology and genesis of brain AVM-a narrative review

Considerable progress has been made over the past years to better understand the genetic nature and pathophysiology of brain AVM. For the actual review, a PubMed search was carried out regarding the embryology, inflammation, advanced imaging, and fluid dynamical modeling of brain AVM. Whole-genome sequencing clarified the genetic origin of sporadic and familial AVM to a large degree, although some open questions remain. Advanced MRI and DSA techniques allow for better segmentation of feeding arteries, nidus, and draining veins, as well as the deduction of hemodynamic parameters such as flow and pressure in the individual AVM compartments. Nonetheless, complete modeling of the intranidal flow structure by computed fluid dynamics (CFD) is not possible so far. Substantial progress has been made towards understanding the embryology of brain AVM. In contrast to arterial aneurysms, complete modeling of the intranidal flow and a thorough understanding of the mechanical properties of the AVM nidus are still lacking at the present time.

Concurrent imaging of vascularization and metabolism in a mouse model of paraganglioma under anti-angiogenic treatment

Rationale: Deregulation of metabolism and induction of vascularization are major hallmarks of cancer. Using a new multimodal preclinical imaging instrument, we explored a sequence of events leading to sunitinib-induced resistance in a murine model of paraganglioma (PGL) invalidated for the expression of succinate dehydrogenase subunit B (Sdhb^-/-).

Methods: Two groups of Sdhb^-/- tumors bearing mice were treated with sunitinib (6 weeks) or vehicle (3 weeks). Concurrent Positron Emission Tomography (PET) with 2′ -deoxy-2′-[¹⁸F]fluoro-D-glucose (FDG), Computed Tomography (CT) and Ultrafast Ultrasound Imaging (UUI) imaging sessions were performed once a week and ex vivo samples were analyzed by western blots and histology.

Results: PET-CT-UUI enabled to detect a rapid growth of Sdhb^-/- tumors with increased glycolysis and vascular development. Sunitinib treatment prevented tumor growth, vessel development and reduced FDG uptake at week 1 and 2 (W1-2). Thereafter, imaging revealed tumor escape from sunitinib treatment: FDG uptake in tumors increased at W3, followed by tumor growth and vessel development at W4-5. Perfused vessels were preferentially distributed in the hypermetabolic regions of the tumors and the perfused volume increased during escape from sunitinib treatment. Finally, initial changes in total lesion glycolysis and maximum vessel length at W1 were predictive of resistance to sunitinib.

Conclusion: These results demonstrate an adaptive resistance of Sdhb^-/- tumors to six weeks of sunitinib treatment. Early metabolic changes and delayed vessel architecture changes were detectable and predictable in vivo early during anti-angiogenic treatment. Simultaneous metabolic, anatomical and functional imaging can monitor precisely the effects of anti-angiogenic treatment of tumors.

Three-dimensional segmentation and symbolic representation of cerebral vessels on 3DRA images of arteriovenous malformations

BACKGROUND

Endovascular embolization is a minimally invasive interventional method for the treatment of neurovascular pathologies such as aneurysms, arterial stenosis or arteriovenous malformations (AVMs). In this context, neuroradiologists need efficient tools for interventional planning and microcatheter embolization procedures optimization. Thus, the development of helpful methods is necessary to solve this challenging issue.

METHODS

A complete pipeline aiming to assist neuroradiologists in the visualization, interpretation and exploitation of three-dimensional rotational angiographic (3DRA) images for interventions planning in case of AVM is proposed. The developed method consists of two steps. First, an automated 3D region-based segmentation of the cerebral vessels which feed and drain the AVM is performed. From this, a graph-like tree representation of these connected vessels is then built. This symbolic representation provides a vascular network modelization with hierarchical and geometrical features that helps in the understanding of the complex angioarchitecture of the AVM.

RESULTS

The developed workflow achieves the segmentation of the vessels and of the malformation. It improves the 3D visualization of this complex network and highlights its three main components that are the arteries, the veins and the nidus. The symbolic representation then brings a better comprehension of the vessels angioarchitecture. It provides decomposition into topologically related vessels, offering the possibility to reduce the complexity due to the malformed vessels and also determine the optimal paths for AVM embolization during interventions planning.

CONCLUSIONS

A relevant vascular network modelization has been developed that constitutes a breakthrough in the assistance of neuroradiologists for AVM endovascular embolization planning.