Automated anatomical labeling of coronary arteries via bidirectional tree LSTMs
Automated anatomical labeling facilitates the diagnostic process for physicians and radiologists. One of the challenges in automated anatomical labeling problems is the robustness to handle the large individual variability inherited in human anatomy. A novel deep neural network framework, referred to Tree Labeling Network (TreeLab-Net), is proposed to resolve this problem in this work.
A multi-layer perceptron (MLP) encoder network and a bidirectional tree-structural long short-term memory (Bi-TreeLSTM) are combined to construct the TreeLab-Net. Vessel spatial locations and directions are selected as features, where a spherical coordinate transform is utilized to normalize vessel spatial variations. The dataset includes 436 coronary computed tomography angiography images. Tenfold cross-validation is performed for evaluation.
The precision–recall curve of TreeLab-Net shows that the four main branch classes, LM, LAD, LCX and RCA, have the area under the curve (AUC) higher than 97%. Other major side branch classes, D, OM, and R-PLB, also have AUC higher than 90%. Comparing with four other methods (i.e., AdaBoost, MLP, Up-to-Down and Down-to-Up TreeLSTM), the TreeLab-Net achieves higher F1 scores with less topological errors.
The TreeLab-Net is able to capture the characteristics of tree structures by learning the spatial and topological dependencies of blood vessels effectively. The results demonstrate that TreeLab-Net is able to yield competitive performances on a large dataset with great variance among subjects.
KeywordsAnatomical labeling Coronary artery Coronary computed tomography angiography Spherical coordinate transform Tree-structural long short-term memory Deep learning
The work received supports from Shenzhen Municipal Government under the Grant KQTD2016112809330877.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
All procedures performed in these studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. For this type of study, formal consent is not required.
- 1.Akinyemi A, Murphy S, Poole I, Roberts C (2009) Automatic labelling of coronary arteries. In: 2009 17th European signal processing conference, pp 1562–1566Google Scholar
- 2.Cao Q, Broersen A, de Graaf MA, Kitslaar PH, Yang G, Scholte AJ, Lelieveldt B, Reiber J, Dijkstra J (2017) Automatic identification of coronary tree anatomy in coronary computed tomography angiography. Int J Cardiovasc Imaging 33:1809–1819. https://doi.org/10.1007/s10554-017-1169-0 CrossRefGoogle Scholar
- 3.Yang G, Broersen A, Petr R, Kitslaar PH, de Graaf MA, Bax JJ, Reiber J, Dijkstra J (2011) Automatic coronary artery tree labeling in coronary computed tomographic angiography datasets. In: 2011 Computing in cardiology, pp 109–112Google Scholar
- 4.Gülsün MA, Funka-Lea G, Zheng Y, Eckert M (2014) CTA coronary labeling through efficient geodesics between trees using anatomy priors. Med Image Comput Comput Assist Interv 17:521–528Google Scholar
- 8.Hoang BH, Oda M, Jiang Z, Kitasaka T, Misawa K, Fujiwara M, Mori K (2011) A study on automated anatomical labeling to arteries concerning with colon from 3D abdominal CT images. In: Medical imaging 2011: image processing. International Society for Optics and Photonics, p 79623RGoogle Scholar
- 11.Zhang W, Liu J, Yao J, Summers RM (2013) Automatic anatomical labeling of abdominal arteries for small bowel evaluation on 3D CT scans. In: 2013 IEEE 10th international symposium on biomedical imaging, pp 210–213Google Scholar
- 14.Tai KS, Socher R, Manning CD (2015) Improved semantic representations from tree-structured long short-term memory networks. arXiv:1503.00075 CsGoogle Scholar
- 15.Graves A, Jaitly N, Mohamed A (2013) Hybrid speech recognition with Deep Bidirectional LSTM. In: 2013 IEEE workshop on automatic speech recognition and understanding, pp 273–278Google Scholar
- 16.Ronneberger O, Fischer P, Brox T (2015) U-Net: convolutional networks for biomedical image segmentation. In: Navab N, Hornegger J, Wells W, Frangi A (eds) Medical image computing and computer-assisted intervention (MICCAI 2015). Lecture notes in computer science, vol 9351. Springer, Cham, pp 234–241Google Scholar