Abstract
Anatomical airway labeling is crucial for precisely identifying airways displaying symptoms such as constriction, increased wall thickness, and modified branching patterns, facilitating the diagnosis and treatment of pulmonary ailments. This study introduces an innovative airway labeling methodology, BranchLabelNet, which accounts for the fractal nature of airways and inherent hierarchical branch nomenclature. In developing this methodology, branch-related parameters, including position vectors, generation levels, branch lengths, areas, perimeters, and more, are extracted from a dataset of 1000 chest computed tomography (CT) images. To effectively manage this intricate branch data, we employ an n-ary tree structure that captures the complicated relationships within the airway tree. Subsequently, we employ a divide-and-group deep learning approach for multi-label classification, streamlining the anatomical airway branch labeling process. Additionally, we address the challenge of class imbalance in the dataset by incorporating the Tomek Links algorithm to maintain model reliability and accuracy. Our proposed airway labeling method provides robust branch designations and achieves an impressive average classification accuracy of 95.94% across fivefold cross-validation. This approach is adaptable for addressing similar complexities in general multi-label classification problems within biomedical systems.
Graphical Abstract
Similar content being viewed by others
References
Park J et al (2020) Subtyping COPD by using visual and quantitative CT imaging features. Chest 157(1):47–60. https://doi.org/10.1016/j.chest.2019.06.015
Dransfield MT, Washko GR, Foreman MG, Estepar RSJ, Reilly J, Bailey WC (2007) Gender differences in the severity of CT emphysema in COPD. Chest 132(2):464–470. https://doi.org/10.1378/chest.07-0863
Ley-Zaporozhan J, Ley S, Kauczor HU (2008) Morphological and functional imaging in COPD with CT and MRI: present and future. Eur Radiol 18(3):510–521. https://doi.org/10.1007/s00330-007-0772-1
Washko GR et al (2009) Airway wall attenuation: a biomarker of airway disease in subjects with COPD. J Appl Physiol 107(1):185–191
Lynch DA (1993) Imaging of small airways diseases. Clin Chest Med 14(4):623–634
Berniker AV, Henry TS (2016) Imaging of small airways diseases. Radiol Clin 54(6):1165–1181
Abbott GF, Rosado-de-Christenson ML, Rossi SE, Suster S (2009) Imaging of small airways disease. J Thorac Imaging 24(4):285–298
"VIDA." https://www.vidalung.ai/. Accessed 26 Apr 2024
C. S. Europe. "AVIEW COPD." https://www.aview-lung.com. Accessed 26 Apr 2024
Pieper S, Halle M, Kikinis R (2004) 3D Slicer. In: 2004 2nd IEEE international symposium on biomedical imaging: nano to macro (IEEE Cat No. 04EX821), IEEE, pp 632–635
Mori K, Hasegawa J-I, Suenaga Y, Toriwaki J-I (2000) Automated anatomical labeling of the bronchial branch and its application to the virtual bronchoscopy system. IEEE Trans Med Imaging 19(2):103–114
Kitaoka H et al (2002) Automated nomenclature labeling of the bronchial tree in 3D-CT lung images. Medical Image Computing and Computer-Assisted Intervention—MICCAI 2002: 5th International Conference Tokyo, Japan, September 25–28, 2002 Proceedings, Part II 5. Springer, pp 1–11
Ross JC et al (2014) Airway labeling using a hidden Markov tree model. In: 2014 IEEE 11th International symposium on biomedical imaging (ISBI), IEEE, pp 554–558
Feragen A et al (2014) Geodesic atlas-based labeling of anatomical trees: application and evaluation on airways extracted from CT. IEEE Trans Med Imaging 34(6):1212–1226
T. Y. Zhao, Z. Z. Yin, J. Wang, D. S. Gao, Y. Q. Chen, and Y. X. Mao (2019) Bronchus segmentation and classification by neural networks and linear programming. Medical image computing and computer assisted intervention - Miccai 2019, Pt Vi, vol. 11769, pp 230-239. https://doi.org/10.1007/978-3-030-32226-7_26
Wang MY et al (2020) Automated labeling of the airway tree in terms of lobes based on deep learning of bifurcation point detection. Med Biol Eng Compu 58(9):2009–2024. https://doi.org/10.1007/s11517-020-02184-y
Nadeem SA, Hoffman EA, Comellas AP, Saha PK (2020) Anatomical labeling of human airway branches using a novel two-step machine learning and hierarchical features. In: Medical imaging 2020: image processing, vol 11313: SPIE, pp 234–240
Xie W, Jacobs C, Charbonnier J-P, van Ginneken B (2022) Structure and position-aware graph neural network for airway labeling. arXiv preprint arXiv:2201.04532
Yu W et al (2022) Tnn: Tree neural network for airway anatomical labeling. IEEE Trans Med Imaging 42(1):103–118
Yu W, Zheng H, Gu Y, Xie F, Sun J, Yang J (2023) AirwayFormer: structure-aware boundary-adaptive transformers for airway anatomical labeling. International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, pp 393–402
Cormen TH, Leiserson CE, Rivest RL, Stein C (2022) Introduction to algorithms. MIT press
Mundra S et al (2022) Classification of imbalanced medical data: an empirical study of machine learning approaches. J Intell Fuzzy Syst 43(2):1933–1946. https://doi.org/10.3233/jifs-219294
Li AJ, Zhang P, M Assoc Comp (2020) Research on unbalanced data processing algorithm base Tomeklinks-Smote. In: Aipr 2020: 2020 3rd International conference on artificial intelligence and pattern recognition, pp 13–17. https://doi.org/10.1145/3430199.3430222
Devi D, Biswas SK, Purkayastha B (2017) Redundancy-driven modified Tomek-link based undersampling: a solution to class imbalance. Pattern Recogn Lett 93:3–12. https://doi.org/10.1016/j.patrec.2016.10.006
Batista GE, Prati RC, Monard MC (2004) A study of the behavior of several methods for balancing machine learning training data. ACM SIGKDD Explorations Newsl 6(1):20–29
Kim T et al (2022) “Quantitative computed tomography imaging-based classification of cement dust-exposed subjects with an artificial neural network technique,” (in English). Comput Biol Med 141:105162–105162. https://doi.org/10.1016/j.compbiomed.2021.105162
Ho TT et al (2021) A 3D-CNN model with CT-based parametric response mapping for classifying COPD subjects. Sci Rep 11(1):34
Park J et al (2021) “Quantitative CT image-based structural and functional changes during asthma acute exacerbations,” (in English). J Appl Physiol 131(3):1056–1066. https://doi.org/10.1152/japplphysiol.00743.2020
Choi S et al (2019) 1D network simulations for evaluating regional flow and pressure distributions in healthy and asthmatic human lungs. J Appl Physiol 127(1):122–133
Tschirren J, Han MK, Barr RG, Hoffman EA (2016) Branching patterns and automated labeling of sub-segmental human airways. In C48. COPD: IMAGING. Am J Respir Crit Care Med vol 193, ISSN: 1073-449X
Lafore R, Broder A, Canning J (2022) Data Structures & Algorithms in Python. Addison-Wesley Professional
Epstein CL (2007) Introduction to the mathematics of medical imaging. SIAM
Epstein CL (2003) Mathematics of medical imaging. ed: Prentice Hall Upper-Saddle River, NJ
Tomek I (1976) Two modifications of CNN. IEEE Transaction on Systems, Man, and Cybernetics. 6(11):769–772. https://doi.org/10.1109/TSMC.1976.4309452
Dietterich TG (2000) Ensemble methods in machine learning. International workshop on multiple classifier systems. Springer, pp 1–15
Schaffer C (1993) Selecting a classification method by cross-validation. Mach Learn 13(1):135–143. https://doi.org/10.1023/a:1022639714137
Hou XL, Guo WC, Ren SJ, Li Y, Si Y, Su LZ (2022) Bolt-loosening detection using 1D and 2D Input data based on two-stream convolutional neural networks. Materials 15(19):6757. https://doi.org/10.3390/ma15196757
Abdoli S, Cardinal P, Koerich AL (2019) End-to-end environmental sound classification using a 1D convolutional neural network. Expert Syst Appl 136:252–263. https://doi.org/10.1016/j.eswa.2019.06.040
Li F et al (2019) Feature extraction and classification of heart sound using 1D convolutional neural networks. Eurasip J Adv Signal Process 2019(1):59. https://doi.org/10.1186/s13634-019-0651-3
Sang XC, Zhou RG, Li YC, Xiong SJ (2022) One-dimensional deep convolutional neural network for mineral classification from Raman spectroscopy. Neural Process Lett 54(1):677–690. https://doi.org/10.1007/s11063-021-10652-1
Xie SL et al (2020) Research on intelligent fault diagnosis method for rolling bearing based on one-dimensional LeNet-5 convolutional neural network. In: 10th Institute-of-Electrical-and-Electronics-Engineers International Conference on Cyber Technology in Automation, Control, and Intelligent Systems (CYBER), Xian, PEOPLES R CHINA, Oct 10–13 2020, in IEEE Annual international conference on cyber technology in automation control and intelligent systems, pp 295–300, https://doi.org/10.1109/cyber50695.2020.9279185. Available: <Go to ISI>://WOS:000646188000051
Glorot X, Bengio Y (2010) Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the thirteenth international conference on artificial intelligence and statistics. JMLR workshop and conference proceedings, pp 249–256
He K, Zhang X, Ren S, Sun J (2015) Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. In: Proceedings of the IEEE international conference on computer vision, pp 1026–1034
Breiman L, Cutler R (2001) Random forests machine learning. J Clin Microbiol 2:199–228
Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP (2002) SMOTE: synthetic minority over-sampling technique. J Artif intell Res 16:321–357
Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556. Accessed 01 Nov 2023
Howard AG et al (2017) Mobilenets: efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861. Accessed 01 Nov 2023
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770–778
Xu K, Hu W, Leskovec J, Jegelka S (2018) How powerful are graph neural networks? arXiv preprint arXiv:1810.00826
Hamilton W, Ying Z, Leskovec J (2017) Inductive representation learning on large graphs. Adv Neural Inf Process Syst, vol. 30. https://api.semanticscholar.org/CorpusID:4755450
Zhang M, Li P (2021) Nested graph neural networks. Adv Neural Inf Process Syst 34:15734–15747
Smith BM et al (2018) Human airway branch variation and chronic obstructive pulmonary disease. Proc Natl Acad Sci 115(5):E974–E981
Zhang M et al (2023) Multi-site, multi-domain airway tree modeling. Med Image Anal 90:102957
Antonelli M et al (2022) The medical segmentation decathlon. Nat Commun 13(1):4128
Funding
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) [grant number NRF-2023R1A2C2003781].
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Glossary
- 1D
-
One-dimensional
- BMI
-
Body mass index
- CNN
-
Convolutional neural network
- COPD
-
Chronic obstructive pulmonary disease
- CT
-
Computed tomography
- D
-
Diameter of a branch
- Dpnode
-
Depth of a node in the n-ary tree
- EToR
-
Euclidean distance from the starting point to a branch in the n-ary tree
- FEV1
-
Forced expiratory volume in one second
- FVC
-
Forced vital capacity
- Gen
-
Generation of a branch
- GIN
-
Graph isomorphic network
- GNNs
-
Graph neural networks
- isInt
-
Is an internal node in the n-ary tree
- isLeaf
-
Is a leaf node in the n-ary tree
- L
-
Length of a branch
- LA
-
Luminal area
- LMB
-
Left main bronchus
- N child
-
Number of children of a node in the n-ary tree
- NGCN
-
Nested graph convolutional network
- N sib
-
Number of siblings of a node in the n-ary tree
- P in
-
Inner perimeter
- P out
-
Outer perimeter
- QCT
-
Quantitative computed tomography
- RMB
-
Right main bronchus
- SAGE
-
GraphSAGE
- SD
-
Standard deviation
- SMOTE
-
Synthetic minority oversampling technique
- SPGNN
-
Structure and position-aware graph neural network
- TA
-
Total area
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Chau, NK., Ma, TT., Kim, W.J. et al. BranchLabelNet: Anatomical Human Airway Labeling Approach using a Dividing-and-Grouping Multi-Label Classification. Med Biol Eng Comput (2024). https://doi.org/10.1007/s11517-024-03119-7
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11517-024-03119-7