Segmentation to Label: Automatic Coronary Artery Labeling from Mask Parcellation

Li, Zhuowei; Xia, Qing; Wang, Wenji; Yan, Zhennan; Yin, Ruohan; Pan, Changjie; Metaxas, Dimitris N.

doi:10.1007/978-3-030-59861-7_14

Cited by 6 publications

(6 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, in line with previous works, instead of the ensemble, only the GCN based on the undivided segments (without a maximum segment length) was used. Columns 3 to 6 list results of previous deep learning-based methods for anatomical labeling, applied to the coronary artery lumen segmentation 22,23 or to the coronary artery centerline tree 15,16 (as our proposed work). Note that training and evaluating our method on the reference trees, as opposed to the automatically extracted trees, enable performance on par with the state of the art.…”

Section: Discussionmentioning

confidence: 99%

“…Proposed labeling based on undivided segments applied to manually corrected tree Li et al 22 Zhang et al 23a Wu et al 15 Yang et al 16 Li et al lead to the ostium, e.g., due to an erroneous choice of tracking direction at a bifurcation. We observed that increasing the number of seed points for tree extraction leads to saturation of recall at a value slightly below 1 (Fig.…”

Section: Discussionmentioning

confidence: 99%

“…15,16 Methods operating on the lumen segmentation did not report graph characteristics. 22,23 A smaller number of subjects and a larger number of segments per tree likely increased the difficulty of anatomical labeling in our work in comparison with previous graph-based methods.…”

Section: Anatomical Labelingmentioning

confidence: 99%

“…This was achieved by analyzing manually segmented coronary artery lumen, i.e., by assigning anatomical labels to lumen voxels using CNNs 22 or by labeling points sampled from the lumen using point cloud-based networks. 22,23 Although analyzing the lumen segmentation enabled learning representations that include information about artery morphology, it has been shown that further processing is required to ensure good performance. Alternatively, machine learning was applied directly to the coronary artery centerline tree.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Graph neural networks for automatic extraction and labeling of the coronary artery tree in CT angiography

Hampe,

van Velzen,

Wolterink

et al. 2024

J. Med. Imag.

View full text Add to dashboard Cite

Automatic comprehensive reporting of coronary artery disease (CAD) requires anatomical localization of the coronary artery pathologies. To address this, we propose a fully automatic method for extraction and anatomical labeling of the coronary artery tree using deep learning.Approach: We include coronary CT angiography (CCTA) scans of 104 patients from two hospitals. Reference annotations of coronary artery tree centerlines and labels of coronary artery segments were assigned to 10 segment classes following the American Heart Association guidelines. Our automatic method first extracts the coronary artery tree from CCTA, automatically placing a large number of seed points and simultaneous tracking of vessel-like structures from these points. Thereafter, the extracted tree is refined to retain coronary arteries only, which are subsequently labeled with a multi-resolution ensemble of graph convolutional neural networks that combine geometrical and image intensity information from adjacent segments. Results:The method is evaluated on its ability to extract the coronary tree and to label its segments, by comparing the automatically derived and the reference labels. A separate assessment of tree extraction yielded an F 1 score of 0.85. Evaluation of our combined method leads to an average F 1 score of 0.74. Conclusions:The results demonstrate that our method enables fully automatic extraction and anatomical labeling of coronary artery trees from CCTA scans. Therefore, it has the potential to facilitate detailed automatic reporting of CAD.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Anatomical Labelingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Graph neural networks for automatic extraction and labeling of the coronary artery tree in CT angiography

Hampe,

van Velzen,

Wolterink

et al. 2024

J. Med. Imag.

View full text Add to dashboard Cite

show abstract

“…After inputting the coding into Point++, they used graph convolutional networks (GCN) to enhance the identification effect of the vessel end. Li [13] converted cardiac chambers and coronary arteries into 3D point clouds and aligned them with each other. V-Net is used to accomplish this task of semantic segmentation.…”

Section: Data-driven Methodsmentioning

confidence: 99%

VP-GAT: vector prior graph attention network for automated segment labeling of coronary arteries

Zhang¹

2023

Fourteenth International Conference on Graphics and Image Processing (ICGIP 2022)

View full text Add to dashboard Cite

Automatic segment labeling of the coronary artery tree is important for computer-aided diagnosis (CAD) of cardiovascular disease. High individual variability among human bodies makes the task very difficult. State-of-the-art methods generally rely on the location information of coronary main branches and image information in a small range, which adversely affects the labeling effect of side branches. We propose a vector prior graph attention network (VP-GAT), which uses image features of organs around the coronary arteries as anatomical prior knowledge, considering the position and direction relationships between segments and surrounding organs. VP-GAT consists of three main parts: image prior GAT, full-vector filed extractor, and image domain prior knowledge extractor. We first extract the anatomical information of the coronary arteries as a full vector field, and then extract the image domain prior knowledge through the hybrid model of ResUnet and Transformer. Finally, we feed the two into the image prior GAT to label the segments. Our method is evaluated on real clinical datasets achieving an F1 score of 95.5%. Extensive experiments show that VP-GAT significantly outperforms state-of-the-art methods in labeling the side branches of coronary arteries.

show abstract

Exhaustive matching of 3D/2D coronary artery structure based on imperfect segmentations

Xie

Zhang

et al. 2023

Int J CARS

View full text Add to dashboard Cite

Segmentation to Label: Automatic Coronary Artery Labeling from Mask Parcellation

Cited by 6 publications

References 11 publications

Graph neural networks for automatic extraction and labeling of the coronary artery tree in CT angiography

Graph neural networks for automatic extraction and labeling of the coronary artery tree in CT angiography

VP-GAT: vector prior graph attention network for automated segment labeling of coronary arteries

Exhaustive matching of 3D/2D coronary artery structure based on imperfect segmentations

Contact Info

Product

Resources

About