DeepVesselNet: Vessel Segmentation, Centerline Prediction, and Bifurcation Detection in 3-D Angiographic Volumes

Tetteh, Giles; Efremov, Velizar; Forkert, Nils D.; Schneider, Matthias; Kirschke, Jan S.; Weber, Bruno; Zimmer, Claus; Piraud, Marie; Menze, Bjoern H.

doi:10.48550/arxiv.1803.09340

Cited by 14 publications

(24 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our graph extraction protocol begins with a given whole-brain vascular segmentation. Independent of segmentation method used (deep learning or filter-based), we tested the following state-of-the-art graph extraction algorithms: 1) the TubeMap method [7] which uses pruning on a 27-neighborhood skeletonization after a deep learning based tube-filling algorithm, based on a modified DeepVesselNet architecture [28]; 2) the metric graph reconstruction algorithm by Aanjaneya et al [29] which reduces linear connections of a skeleton to form a more compact and topologically correct graph and 3) the Voreen vessel graph extraction method [30,31]. We tested the graph extraction algorithms on different imaging modalities, varying brain areas, and synthetically generated vascular trees [32].…”

Section: Graph Extraction From Segmentationmentioning

confidence: 99%

Whole Brain Vessel Graphs: A Dataset and Benchmark for Graph Learning and Neuroscience (VesselGraph)

Paetzold,

McGinnis,

Shit

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Biological neural networks define the brain function and intelligence of humans and other mammals, and form ultra-large, spatial, structured graphs. Their neuronal organization is closely interconnected with the spatial organization of the brain's microvasculature, which supplies oxygen to the neurons and builds a complementary spatial graph. This vasculature (or the vessel structure) plays an important role in neuroscience; for example, the organization of (and changes to) vessel structure can represent early signs of various pathologies, e.g. Alzheimer's disease or stroke. Recently, advances in tissue clearing have enabled whole brain imaging and segmentation of the entirety of the mouse brain's vasculature. Building on these advances in imaging, we are presenting an extendable dataset of whole-brain vessel graphs based on specific imaging protocols. Specifically, we extract vascular graphs using a refined graph extraction scheme leveraging the volume rendering engine Voreen and provide them in an accessible and adaptable form through the OGB and PyTorch Geometric dataloaders. Moreover, we benchmark numerous state-of-the-art graph learning algorithms on the biologically relevant tasks of vessel prediction and vessel classification using the introduced vessel graph dataset. Our work paves a path towards advancing graph learning research into the field of neuroscience. Complementarily, the presented dataset raises challenging graph learning research questions for the machine learning community, in terms of incorporating biological priors into learning algorithms, or in scaling these algorithms to handle sparse,spatial graphs with millions of nodes and edges. 1 1 All datasets and code are available for download at https://github.com/jocpae/VesselGraph.Preprint. Under review.

show abstract

Section: Graph Extraction From Segmentationmentioning

confidence: 99%

Whole Brain Vessel Graphs: A Dataset and Benchmark for Graph Learning and Neuroscience (VesselGraph)

Paetzold,

McGinnis,

Shit

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In stark contrast to previous works, where segmentation and centerline prediction has been learned jointly as multi-task learning [37,34], we are not interested in learning the centerline. We are interested in learning a topologypreserving segmentation.…”

Section: Missing Skeletonmentioning

confidence: 99%

clDice -- a Novel Topology-Preserving Loss Function for Tubular Structure Segmentation

Shit,

Paetzold,

Sekuboyina

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Accurate segmentation of tubular, network-like structures, such as vessels, neurons, or roads, is relevant to many fields of research. For such structures, the topology is their most important characteristic, e.g. preserving connectedness: in case of vascular networks, missing a connected vessel entirely alters the blood-flow dynamics. We introduce a novel similarity measure termed clDice, which is calculated on the intersection of the segmentation masks and their (morphological) skeletons. Crucially, we theoretically prove that clDice guarantees topological correctness for binary 2D and 3D segmentation. Extending this, we propose a computationally efficient, differentiable soft-clDice as a loss function for training arbitrary neural segmentation networks. We benchmark the soft-clDice loss for segmentation on four public datasets (2D and 3D). Training on soft-clDice leads to segmentation with more accurate connectivity information, higher graph similarity, and better volumetric scores.

show abstract

“…For vessel segmentation task, the object of interest accounts far less than the background voxels in most cases, which leads to a high rate of false positive and recall values [12]. To alleviate the class-imbalance problem caused by the inequitable penalty of positive and negative voxels, we separate the training process as two stage: 1) In the pre-trained stage, we use NLL(Negative Log Likelihood) loss to get a coarse model; 2) In the fine-tuned stage, we resume the coarse model and adopt a weight-balanced loss to suppress the over-segmentation and high false-positive rate: the calculated voxel-wised losses of both positive and negative positions are sorted, and the negative sorted list is much longer than the positive one considering the small occupancy of interest regions.…”

Section: Two-stage Loss Functionmentioning

confidence: 99%

Pulmonary Vessel Segmentation Based on Orthogonal Fused U-Net++ of Chest CT Images

Cui

Liu

Huang

2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Pulmonary vessel segmentation is important for clinical diagnosis of pulmonary diseases, while is also challenging due to the complicated structure. In this work, we present an effective framework and refinement process of pulmonary vessel segmentation from chest computed tomographic (CT) images. The key to our approach is a 2.5D segmentation network applied from three orthogonal axes, which presents a robust and fully automated pulmonary vessel segmentation result with lower network complexity and memory usage compared to 3D networks. The slice radius is introduced to convolve the adjacent information of the center slice and the multi-planar fusion optimizes the presentation of intra and inter slice features. Besides, the tree-like structure of pulmonary vessel is extracted in the post-processing process, which is used for segmentation refining and pruning. In the evaluation experiments, three fusion methods are tested and the most promising one is compared with the state-of-the-art 2D and 3D structures on 300 cases of lung images randomly selected from LIDC dataset. Our method outperforms other network structures by a large margin and achieves by far the highest average DICE score of 0.9272 and a precision of 0.9310, as per our knowledge from the pulmonary vessel segmentation models available in literature.

show abstract

DeepVesselNet: Vessel Segmentation, Centerline Prediction, and Bifurcation Detection in 3-D Angiographic Volumes

Cited by 14 publications

References 0 publications

Whole Brain Vessel Graphs: A Dataset and Benchmark for Graph Learning and Neuroscience (VesselGraph)

Whole Brain Vessel Graphs: A Dataset and Benchmark for Graph Learning and Neuroscience (VesselGraph)

clDice -- a Novel Topology-Preserving Loss Function for Tubular Structure Segmentation

Pulmonary Vessel Segmentation Based on Orthogonal Fused U-Net++ of Chest CT Images

Contact Info

Product

Resources

About