Skip to main content

Automated segmentation and quantification of the healthy and diseased aorta in CT angiographies using a dedicated deep learning approach

Abstract

Objectives

To develop and validate a deep learning–based algorithm for segmenting and quantifying the physiological and diseased aorta in computed tomography angiographies.

Methods

CTA exams of the aorta of 191 patients (68.1 ± 14 years, 128 male), performed between 2015 and 2018, were retrospectively identified from our imaging archive and manually segmented by two investigators. A 3D U-Net model was trained on the data, which was divided into a training, a validation, and a test group at a ratio of 7:1:2. Cases in the test group (n = 41) were evaluated to compare manual and automatic segmentations. Dice similarity coefficient (DSC), mean surface distance (MSD), and Hausdorff surface distance (HSD) were extracted. Maximum diameter, effective diameter, and area were quantified and compared between both segmentations at eight anatomical landmarks, and at the maximum area of an aneurysms if present (n = 14). Statistics included error calculation, intraclass correlation coefficient, and Bland-Altman analysis.

Results

A DSC of 0.95 [0.94; 0.95] and an MSD of 0.76 [0.06; 0.99] indicated close agreement between segmentations. HSD was 8.00 [4.47; 10.00]. The largest absolute errors were found in the ascending aorta with 0.8 ± 1.5 mm for maximum diameter and at the coeliac trunk with − 30.0 ± 81.6 mm2 for area. Results for absolute errors in aneurysms were − 0.5 ± 2.3 mm for maximum diameter, 0.3 ± 1.6 mm for effective diameter, and 64.9 ± 114.9 mm2 for area. ICC showed excellent agreement (> 0.9; p < 0.05) between quantitative measurements.

Conclusions

Automated segmentation of the aorta on CTA data using a deep learning algorithm is feasible and allows for accurate quantification of the aortic lumen even if the vascular architecture is altered by disease.

Key Points

A deep learning–based algorithm can automatically segment the aorta, mostly within acceptable margins of error, even if the vascular architecture is altered by disease.

Quantifications performed in the segmentations were mostly within clinically acceptable limits, even in pathologically altered segments of the aorta.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Abbreviations

3D:

Three-dimensional

CNN:

Convolutional neural networks

DSC:

Dice similarity coefficient

Eff. diameter:

Effective vessel diameter

HSD:

Hausdorff surface distance

ICC:

Intraclass correlation coefficient

Max. diameter:

Maximum vessel diameter

MSD:

Mean surface distance

References

  1. Roth GA, Huffman MD, Moran AE et al (2015) Global and regional patterns in cardiovascular mortality from 1990 to 2013. Circulation 132:1667–1678

    Article  Google Scholar 

  2. Sakalihasan N, Limet R, Defawe OD (2005) Abdominal aortic aneurysm. Lancet 365:1577–1589

    CAS  Article  Google Scholar 

  3. Iezzi R, Goldberg SN, Merlino B, Posa A, Valentini V, Manfredi R (2019) Artificial intelligence in interventional radiology: a literature review and future perspectives. J Oncol 2019:6153041

    Article  Google Scholar 

  4. Moccia S, De Momi E, El Hadji S, Mattos LS (2018) Blood vessel segmentation algorithms - review of methods, datasets and evaluation metrics. Comput Methods Programs Biomed 158:71–91

    Article  Google Scholar 

  5. Kamman AV, van Herwaarden JA, Orrico M et al (2016) Standardized protocol to analyze computed tomography imaging of type B aortic dissections. J Endovasc Ther 23:472–482

    Article  Google Scholar 

  6. Hosny A, Parmar C, Quackenbush J, Schwartz LH, Aerts HJWL (2018) Artificial intelligence in radiology. Nat Rev Cancer 18:500–510

    CAS  Article  Google Scholar 

  7. Cao L, Shi R, Ge Y et al (2019) Fully automatic segmentation of type B aortic dissection from CTA images enabled by deep learning. Eur J Radiol 121:108713

    Article  Google Scholar 

  8. Lopez-Linares K, Aranjuelo N, Kabongo L et al (2018) Fully automatic detection and segmentation of abdominal aortic thrombus in post-operative CTA images using deep convolutional neural networks. Med Image Anal 46:202–214

    Article  Google Scholar 

  9. Sedghi Gamechi Z, Bons LR, Giordano M et al (2019) Automated 3D segmentation and diameter measurement of the thoracic aorta on non-contrast enhanced CT. Eur Radiol 29:4613–4623

    Article  Google Scholar 

  10. Zlahoda-Huzior A, Stanuch M, Witowski J, Dudek D (2019) Automatic aorta and left ventricle segmentation for TAVI procedure planning using convolutional neural networks. Conference Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society 2019:2777–2780

  11. Hahn LD, Mistelbauer G, Higashigaito K et al (2020) CT-based true- and false-lumen segmentation in type B aortic dissection using machine learning. Radiol Cardiothorac Imaging 2:e190179

  12. Yu Y, Gao Y, Wei J et al (2021) A three-dimensional deep convolutional neural network for automatic segmentation and diameter measurement of type B aortic dissection. Korean J Radiol 22:168–178

    Article  Google Scholar 

  13. Klein J, Wenzel M, Romberg D et al (2020) QuantMed: component-based deep learning platform for translational research. SPIE

  14. Herman GT, Zheng J, Bucholtz CA (1992) Shape-based interpolation. IEEE Comput Graph Appl 12:69–79

    Article  Google Scholar 

  15. Çiçek Ö, Abdulkadir A, Lienkamp SS, Brox T, Ronneberger O (2016) 3D U-Net: learning dense volumetric segmentation from sparse annotation. arXiv e-prints

  16. Dice LR (1945) Measures of the amount of ecologic association between species. Ecology 26:297–302

    Article  Google Scholar 

  17. Gerig G, Jomier M, Chakos M (2001) Valmet: a new validation tool for assessing and improving 3D object segmentationproceedings of the 4th International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer-Verlag, pp 516–523

  18. Cardenes R, de Luis-Garcia R, Bach-Cuadra M (2009) A multidimensional segmentation evaluation for medical image data. Comput Methods Programs Biomed 96:108–124

    Article  Google Scholar 

  19. Selle D, Preim B, Schenk A, Peitgen H (2002) Analysis of vasculature for liver surgical planning. IEEE Trans Med Imaging 21:1344–1357

    Article  Google Scholar 

  20. Koo TK, Li MY (2016) A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med 15:155–163

    Article  Google Scholar 

  21. Krissian K, Malandain G, Ayache N (1997) Directional anisotropic diffusion applied to segmentation of vessels in 3D images. Springer, Berlin Heidelberg, pp 345–348

    Google Scholar 

  22. Frangi AF, Niessen WJ, Vincken KL, Viergever MA (1998) Multiscale vessel enhancement filtering. Springer, Berlin Heidelberg, pp 130–137

    Google Scholar 

  23. Wörz S, Tengg-Kobligk H, Henninger V et al (2010) 3-D quantification of the aortic arch morphology in 3-D CTA data for endovascular aortic repair. IEEE Trans Biomed Eng 57:2359–2368

    Article  Google Scholar 

  24. Biesdorf A, Wörz S, Tengg-Kobligk Hv, Rohr K, Schnörr C (2015) 3D segmentation of vessels by incremental implicit polynomial fitting and convex optimization2015 IEEE 12th International Symposium on Biomedical Imaging (ISBI), pp 1540–1543

  25. Lee SH, Lee S (2015) Adaptive Kalman snake for semi-autonomous 3D vessel tracking. Comput Methods Programs Biomed 122:56–75

    Article  Google Scholar 

  26. Biesdorf A, Rohr K, Feng D et al (2012) Segmentation and quantification of the aortic arch using joint 3D model-based segmentation and elastic image registration. Med Image Anal 16:1187–1201

    Article  Google Scholar 

  27. Ecabert O, Peters J, Walker MJ et al (2011) Segmentation of the heart and great vessels in CT images using a model-based adaptation framework. Med Image Anal 15:863–876

    Article  Google Scholar 

  28. Tian Y, Chen Q, Wang W et al (2014) A vessel active contour model for vascular segmentation. Biomed Res Int 2014:106490

    PubMed  PubMed Central  Google Scholar 

  29. Quint LE, Liu PS, Booher AM, Watcharotone K, Myles JD (2013) Proximal thoracic aortic diameter measurements at CT: repeatability and reproducibility according to measurement method. Int J Cardiovasc Imaging 29:479–488

    Article  Google Scholar 

  30. Lalys F, Esneault S, Castro M et al (2019) Automatic aortic root segmentation and anatomical landmarks detection for TAVI procedure planning. Minim Invasive Ther Allied Technol 28:157–164

    Article  Google Scholar 

  31. Gao X, Boccalini S, Kitslaar PH et al (2019) A novel software tool for semi-automatic quantification of thoracic aorta dilatation on baseline and follow-up computed tomography angiography. Int J Cardiovasc Imaging 35:711–723

    Article  Google Scholar 

  32. Brendan McMahan H, Moore E, Ramage D, Hampson S, Agüera y Arcas B (2016) Communication-efficient learning of deep networks from decentralized data. arXiv e-prints

  33. Shokri R, Shmatikov V (2015) Privacy-Preserving Deep Learning Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security. Association for Computing Machinery, Denver, Colorado, USA, pp 1310–1321

Download references

Acknowledgements

We deeply thank Dr. Karina Dietermann for her assistance in the statistical analysis.

Funding

The authors state that this work has not received any funding.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Malte Maria Sieren.

Ethics declarations

Guarantor

The scientific guarantor of this publication is Malte Maria Sieren.

Conflict of interest

The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article.

Statistics and biometry

Dr. Karina Dieterman kindly provided statistical advice for this manuscript.

Informed consent

Written informed consent was waived by the Institutional Review Board.

All patients were consented to the clinically indicated CT examination per clinical standard.

Ethical approval

Institutional Review Board approval was obtained.

Methodology

• retrospective

• experimental/diagnostic study

• performed at one institution

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Sieren, M.M., Widmann, C., Weiss, N. et al. Automated segmentation and quantification of the healthy and diseased aorta in CT angiographies using a dedicated deep learning approach. Eur Radiol 32, 690–701 (2022). https://doi.org/10.1007/s00330-021-08130-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00330-021-08130-2

Keywords

  • Artificial intelligence
  • Neuronal networks
  • Image processing
  • Computer-assisted, Aorta