Abstract
We propose and develop an end-to-end learning pipeline incorporating lung tissue appearance and deformation pattern during two phase respiratory CT chest imaging. We formulate the task as a voxel level classification and quantification problem for the emphysematous areas of lung parenchyma on CT imaging and develop a two-stream fully convolutional neural network which fuses together lung tissue pattern and deformation map in a unified framework. The Jacobian matrix, computed between inspiratory and expiratory images, is used as a surrogate measurement of lung deformation. The DenseVNet architecture is adopted for memory efficiency and accuracy on smaller structure segmentation.
The proposed model is trained from inspiratory and expiratory CT images, where the Jacobian matrix calculates the volumetric deformation map of the lung during respiration from inspiration to expiration: the former is used for tissue appearance of the lung while both together provide motion characteristics. Our ablation study shows that incorporating tissue appearance and lung motion accurately classifies all patients with different types and severity of emphysema against the non-emphysema cases. In addition, the proposed method outperforms previous state-of-the-art methods in grading and severity calculation of emphysema, which enables automatic image-based emphysema grading for the first time. To the best of our knowledge, our work is the first proposed framework to incorporate both tissue appearance and deformation map in a deep learning platform for disease diagnosis in lung using CT imaging, which can be intuitively extended to other moving organs and diseases.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Fernandes, L., Gulati, N., Mesquita, A.M., Sardesai, M., Fernandes, Y.: Quantification of emphysema in chronic obstructive pulmonary disease by volumetric computed tomography of lung. Indian J. Chest Dis. Allied Sci. 57, 155–160 (2015)
Omori, H., Fujimoto, K., Katoh, T.: Computed-tomography findings of emphysema: correlation with spirometric values. Curr. Opin. Pulm. Med. 14, 110–114 (2008)
Wille, M.M.W., et al.: Visual assessment of early emphysema and interstitial abnormalities on CT is useful in lung cancer risk analysis. Eur. Radiol. 26(2), 487–494 (2015). https://doi.org/10.1007/s00330-015-3826-9
Lynch, D.A., Austin, J.H., Hogg, J.C., Grenier, P.A., Kauczor, H.U., et al.: CT-definable subtypes of chronic obstructive pulmonary disease: a statement of the fleischner society. Radiology 277, 192–205 (2015)
Hohberger, L.A., Schroeder, D.R., Bartholmai, B.J., Yang, P., Wendt, C.H., et al.: Correlation of regional emphysema and lung cancer: a lung tissue research consortium-based study. J. Thorac. Oncol. 9, 639–645 (2014)
Haruna, A., Muro, S., Nakano, Y., Ohara, T., Hoshino, Y., et al.: CT scan findings of emphysema predict mortality in COPD. Chest 138, 635–640 (2010)
Anthimopoulos, M., Christodoulidis, S., Ebner, L., Christe, A., Mougiakakou, S.: Lung pattern classification for interstitial lung diseases using a deep convolutional neural network. IEEE Trans. Med. Imaging 35, 1207–1216 (2016)
Gao, M., Bagci, U., Lu, L., Wu, A., Buty, M., et al.: Holistic classification of CT attenuation patterns for interstitial lung diseases via deep convolutional neural networks. Comput. Methods Biomech. Biomed. Eng. Imaging Visual. 6, 1–6 (2018)
Ørting, S. N., Petersen, J., Thomsen, L. H., Wille, M. M. W., and Bruijne, M. d.: Detecting emphysema with multiple instance learning. In: 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), pp. 510–513. (2018)
Christodoulidis, S., Anthimopoulos, M., Ebner, L., Christe, A., Mougiakakou, S.: Multisource transfer learning with convolutional neural networks for lung pattern analysis. IEEE J. Biomed. Health Inform. 21, 76–84 (2017)
Negahdar, M., Beymer, D.: Lung tissue characterization for emphysema differential diagnosis using deep convolutional neural networks. In: SPIE Medical Imaging, vol. 10950, pp. 109503R. San Diego, CA (2019)
Bermejo-Peláez, D., Estepar, R.S.J., Ledesma-Carbayo, M.J.: Emphysema classification using a multi-view convolutional network. In: 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), pp. 519–522 (2018)
Fischer, A.M., Varga-Szemes, A., van Assen, M., Griffith, L.P., Sahbaee, P. et al.: Comparison of artificial intelligence–based fully automatic chest CT emphysema quantification to pulmonary function testing. Am. J. Roentgenol. 214(5), 1065–1071 (2020)
Kauczor, H.-U., Wielpütz, M.O., Jobst, B.J., Weinheimer, O., Gompelmann, D., et al.: Computed tomography imaging for novel therapies of chronic obstructive pulmonary disease. J. Thorac. Imaging 34, 202–213 (2019)
Negahdar, M., Amini, A.A.: Regional lung strains via a volumetric mass conserving optical flow model. In: IEEE International Symposium on Biomedical Imaging (ISBI): From Nano to Macro, pp. 1475–1478. Barcelona, Spain (2012)
Negahdar, M., Dunlap, N., Zacarias, A., Civelek, A.C., Woo, S.Y. et al.: Comparison of indices of regional lung function from 4-D X-ray CT: Jacobian vs. strain of deformation. In: IEEE International Symposium on Biomedical Imaging (ISBI): From Nano to Macro, pp. 648–651. San Francisco, CA, USA (2013)
Negahdar, M., Fasola, C.E., Yu, A.S., von Eyben, R., Yamamoto, T., et al.: Noninvasive pulmonary nodule elastometry by CT and deformable image registration. Radiother. Oncol. 115, 35–40 (2015)
Joe Yue-Hei, N., Hausknecht, M., Vijayanarasimhan, S., Vinyals, O., Monga, R. et al.: Beyond short snippets: deep networks for video classification. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4694–4702 (2015)
Wang, L., Ouyang, W., Wang, X., Lu, H.: Visual tracking with fully convolutional networks. In: 2015 IEEE International Conference on Computer Vision (ICCV), pp. 3119–3127 (2015)
Jain, S.D., Xiong, B., Grauman, K.: FusionSeg: learning to combine motion and appearance for fully automatic segmentation of generic objects in videos. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2117–2126 (2017)
Li, S., Seybold, B., Vorobyov, A., Lei, X., Kuo, C.C.J.: Unsupervised video object segmentation with motion-based bilateral networks. In: Ferrari, V., Hebert, M., Sminchisescu, C., et al. (eds.) Computer Vision – ECCV 2018, pp. 215-231. Cham (2018)
Gibson, E., Giganti, F., Hu, Y., Bonmati, E., Bandula, S., et al.: Towards image-guided pancreas and biliary endoscopy: automatic multi-organ segmentation on abdominal CT with dense dilated networks. In: Medical Image Computing and Computer Assisted Intervention − MICCAI 2017, pp. 728-736. Cham (2017)
Milletari, F., Navab, N., Ahmadi, S.-A.: V-net: fully convolutional neural networks for volumetric medical image segmentation. In: 3D Vision (3DV), 2016 Fourth International Conference on, pp. 565–571 (2016)
Negahdar, M., Beymer, D., Syeda-Mahmood, T.: Automated volumetric lung segmentation of thoracic CT images using fully convolutional neural network. In: SPIE Medical Imaging, vol. 10575 (2018)
Gibson, E., Giganti, F., Hu, Y., Bonmati, E., Bandula, S., et al.: Automatic multi-organ segmentation on abdominal CT with dense V-networks. IEEE Trans. Med. Imaging 37, 1822–1834 (2018)
Ghavami, N., Hu, Y., Gibson, E., Bonmati, E., Emberton, M., et al.: Automatic segmentation of prostate MRI using convolutional neural networks: investigating the impact of network architecture on the accuracy of volume measurement and MRI-ultrasound registration. Med. Image Anal. 58, 101558 (2019)
Klein, S., Staring, M., Murphy, K., Viergever, M.A., Pluim, J.: elastix: a toolbox for intensity-based medical image registration. IEEE Trans. Med. Imaging 29, 196–205 (2010)
Kalogeiton, V., Weinzaepfel, P., Ferrari, V., Schmid, C.: Action tubelet detector for spatio-temporal action localization. In: 2017 IEEE International Conference on Computer Vision (ICCV), pp. 4415–4423 (2017)
Singh, G., Saha, S., Sapienza, M., Torr, P., Cuzzolin, F.: Online real-time multiple spatiotemporal action localisation and prediction. In: 2017 IEEE International Conference on Computer Vision (ICCV), pp. 3657–3666 (2017)
Mohamed Hoesein, F.A., van Rikxoort, E., van Ginneken, B., de Jong, P.A., Prokop, M., et al.: Computed tomography-quantified emphysema distribution is associated with lung function decline. Eur. Respir. J. 40, 844–850 (2012)
Bailey, K.L.: The importance of the assessment of pulmonary function in COPD. Med. Clin. North Am. 96, 745–752 (2012)
Wille, M.M.W., Thomsen, L.H., Dirksen, A., Petersen, J., Pedersen, J.H., Shaker, S.B.: Emphysema progression is visually detectable in low-dose CT in continuous but not in former smokers. Eur. Radiol. 24(11), 2692–2699 (2014). https://doi.org/10.1007/s00330-014-3294-7
Heussel, C.P., Herth, F.J.F., Kappes, J., Hantusch, R., Hartlieb, S., et al.: Fully automatic quantitative assessment of emphysema in computed tomography: comparison with pulmonary function testing and normal values. Eur. Radiol. 19, 2391–2402 (2009)
Bartholmai, B., Karwoski, R., Zavaletta, V., Robb, R., Holmes, D.: The lung tissue research consortium: an extensive open database containing histological, clinical, and radiological data to study chronic lung disease. In: The Insight Journal—2006 MICCAI Open Science Workshop (2006)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Negahdar, M. (2022). Automatic Grading of Emphysema by Combining 3D Lung Tissue Appearance and Deformation Map Using a Two-Stream Fully Convolutional Neural Network. In: Lian, C., Cao, X., Rekik, I., Xu, X., Cui, Z. (eds) Machine Learning in Medical Imaging. MLMI 2022. Lecture Notes in Computer Science, vol 13583. Springer, Cham. https://doi.org/10.1007/978-3-031-21014-3_19
Download citation
DOI: https://doi.org/10.1007/978-3-031-21014-3_19
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-21013-6
Online ISBN: 978-3-031-21014-3
eBook Packages: Computer ScienceComputer Science (R0)