Auto-segmentation; carotid MRI; U-Net; uncertainty regularization; vessel segmentation; Auto segmentation; Biomedical imaging; Blood; Carotid artery; Carotid MRI; Regularisation; Three-dimensional display; U-net; Uncertainty; Uncertainty regularization; Vessel segmentation; Computer Science (all); Materials Science (all); Engineering (all); Electrical and Electronic Engineering; General Engineering; General Materials Science; General Computer Science
Abstract :
[en] We present a fully automated method for carotid artery (CA) outer wall segmentation in black blood MRI using partially annotated data and compare it to the state-of-the-art reference model. Our model was trained and tested on multicentric data of patients (106 and 23 patients, respectively) with a carotid plaque and was validated on different MR sequences (24 patients) as well as data that were acquired with MRI systems of a different vendor (34 patients). A 3D nnU-Net was trained on pre-contrast T1w turbo spin echo (TSE) MR images. A CA centerline sliding window approach was chosen to refine the nnU-Net segmentation using an additionally trained 2D U-Net to increase agreement with manual annotations. To improve segmentation performance in areas with semantically and visually challenging voxels, Monte-Carlo dropout was used. To increase generalizability, data were augmented with intensity transformations. Our method achieves state-of-the-art results yielding a Dice similarity coefficient (DSC) of 91.7% (interquartile range (IQR) 3.3%) and volumetric intraclass correlation (ICC) with ground truth of 0.90 on the development domain data and a DSC of 91.1% (IQR 7.2%) and volumetric ICC with ground truth of 0.83 on the external domain data outperforming top-ranked methods for open-source CA segmentation. The uncertainty-based approach increases the interpretability of the proposed method by providing an uncertainty map together with the segmentation.
Disciplines :
Computer science
Author, co-author :
Lavrova, Elizaveta ; Université de Liège - ULiège > GIGA ; GROW-School for Oncology and Reproduction, Maastricht University, D-Laboratory, Department of Precision Medicine, Maastricht, Netherlands
Salahuddin, Zohaib; GROW-School for Oncology and Reproduction, Maastricht University, D-Laboratory, Department of Precision Medicine, Maastricht, Netherlands
Woodruff, Henry C. ; GROW-School for Oncology and Reproduction, Maastricht University, D-Laboratory, Department of Precision Medicine, Maastricht, Netherlands ; Maastricht University Medical Centre+, Department of Radiology and Nuclear Medicine, Maastricht, Netherlands
Kassem, Mohamed; Maastricht University Medical Centre+, Department of Radiology and Nuclear Medicine, Maastricht, Netherlands ; Maastricht University, CARIM School for Cardiovascular Diseases, Maastricht, Netherlands
Camarasa, Robin; Erasmus University Medical Center Rotterdam, Department of Radiology and Nuclear Medicine, Rotterdam, Netherlands
Van Kolk, Anja G. Der ; University Medical Center Utrecht, Department of Radiology, Utrecht, Netherlands ; Radboudumc, Department of Medical Imaging, Nijmegen, Netherlands
Nederkoorn, Paul J.; Amsterdam UMC, Department of Neurology, Amsterdam, Netherlands
Bos, Daniel; Erasmus University Medical Center Rotterdam, Department of Radiology and Nuclear Medicine, Rotterdam, Netherlands
Hendrikse, Jeroen; University Medical Center Utrecht, Department of Radiology, Utrecht, Netherlands
Kooi, M. Eline ; Maastricht University Medical Centre+, Department of Radiology and Nuclear Medicine, Maastricht, Netherlands ; Maastricht University, CARIM School for Cardiovascular Diseases, Maastricht, Netherlands
Lambin, Philippe; GROW-School for Oncology and Reproduction, Maastricht University, D-Laboratory, Department of Precision Medicine, Maastricht, Netherlands ; Maastricht University Medical Centre+, Department of Radiology and Nuclear Medicine, Maastricht, Netherlands
Language :
English
Title :
UR-CarA-Net: A Cascaded Framework with Uncertainty Regularization for Automated Segmentation of Carotid Arteries on Black Blood MR Images
Publication date :
2023
Journal title :
IEEE Access
ISSN :
2169-3536
Publisher :
Institute of Electrical and Electronics Engineers Inc.
H2020 - 952172 - CHAIMELEON - Accelerating the lab to market transition of AI tools for cancer management
Funders :
EU - European Union [BE] IMI - Innovative Medicines Initiative [BE]
Funding text :
This work was supported in part by the European Union’s Horizon 2020 Research and Innovation Program CHAIMELEON under Grant
952172, in part by EuCanImage under Grant 952103, in part by the TRANSCAN Joint Transnational Call 2016 (JTC2016 CLEARLY)
under Grant UM 2017-8295, and in part by the Innovative Medicines Initiative (IMI)-Optimal Treatment for Patients with Solid Tumours
in Europe Through Artificial intelligence (OPTIMA) under Grant 101034347. The work of Elizaveta Lavrova was supported by the
Liege-Maastricht Imaging Valley Grant.
E. J. Benjamin et al., "Heart disease and stroke statistics-2019 update: A report from the American Heart Association," Circulation, vol. 139, no. 10, pp. e56-e528, Mar. 2019.
M. T. B. Truijman, M. E. Kooi, A. C. van Dijk, A. A. J. de Rotte, A. G. van der Kolk, M. I. Liem, F. H. B. M. Schreuder, E. Boersma, W. H. Mess, R. J. van Oostenbrugge, P. J. Koudstaal, L. J. Kappelle, P. J. Nederkoorn, A. J. Nederveen, J. Hendrikse, A. F. W. van der Steen, M. J. A. P. Daemen, and A. van der Lugt, "Plaque at RISK (PARISK): Prospective multicenter study to improve diagnosis of high-risk carotid plaques," Int. J. Stroke, vol. 9, no. 6, pp. 747-754, Aug. 2014.
K. P. H. Nies, L. J. M. Smits, M. Kassem, P. J. Nederkoorn, R. J. V. Oostenbrugge, and M. E. Kooi, "Emerging role of carotid MRI for personalized ischemic stroke risk prediction in patients with carotid artery stenosis," Frontiers Neurol., vol. 12, Aug. 2021, Art. no. 718438.
A. C. van Dijk, M. T. B. Truijman, B. Hussain, T. Zadi, G. Saiedie, A. A. J. de Rotte, M. I. Liem, A. F. W. van der Steen, M. J. A. P. Daemen, P. J. Koudstaal, P. J. Nederkoorn, J. Hendrikse, M. E. Kooi, and A. van der Lugt, "Intraplaque hemorrhage and the plaque surface in carotid atherosclerosis: The plaque at RISK study (PARISK)," Amer. J. Neuroradiol., vol. 36, no. 11, pp. 2127-2133, Nov. 2015.
L. Saba, C. Yuan, T. S. Hatsukami, N. Balu, Y. Qiao, J. K. DeMarco, T. Saam, A. R. Moody, D. Li, C. C. Matouk, M. H. Johnson, H. R. J ger, M. Mossa-Basha, M. E. Kooi, Z. Fan, D. Saloner, M. Wintermark, D. J. Mikulis, and B. A. Wasserman, "Carotid artery wall imaging: Perspective and guidelines from the ASNR vessel wall imaging study group and expert consensus recommendations of the American society of neuroradiology," Amer. J. Neuroradiol., vol. 39, no. 2, pp. E9-E31, Feb. 2018.
D. H. van Dam-Nolen et al., "Carotid plaque characteristics predict recurrent ischemic stroke and TIA: The PARISK (plaque at risk) study," Cardiovascular Imag., vol. 15, no. 10, pp. 1715-1726, 2022.
A. Saxena, E. Y. K. Ng, and S. T. Lim, "Imaging modalities to diagnose carotid artery stenosis: Progress and prospect," BioMed. Eng. OnLine, vol. 18, May 2019, Art. no. 66.
A. van Engelen, A. C. van Dijk, M. T. B. Truijman, R. Van't Klooster, A. van Opbroek, A. van der Lugt, W. J. Niessen, M. E. Kooi, and M. de Bruijne, "Multi-center MRI carotid plaque component segmentation using feature normalization and transfer learning," IEEE Trans. Med. Imag., vol. 34, no. 6, pp. 1294-1305, Jun. 2015.
I. M. Adame, R. J. van der Geest, B. A. Wasserman, M. A. Mohamed, J. H. C. Reiber, and B. P. F. Lelieveldt, "Automatic segmentation and plaque characterization in atherosclerotic carotid artery MR images," Magn. Reson. Mater. Phys., Biol. Med., vol. 16, no. 5, pp. 227-234, Apr. 2004.
D. S. Jodas, A. S. Pereira, and J. M. R. S. Tavares, "Lumen segmentation in magnetic resonance images of the carotid artery," Comput. Biol. Med., vol. 79, pp. 233-242, Dec. 2016.
A. M. Arias-Lorza, J. Petersen, A. van Engelen, M. Selwaness, A. van der Lugt,W. J. Niessen, and M. de Bruijne, "Carotid artery wall segmentation in multispectral MRI by coupled optimal surface graph cuts," IEEE Trans. Med. Imag., vol. 35, no. 3, pp. 901-911, Mar. 2016.
S. Gao, R. van't Klooster, P. H. Kitslaar, B. F. Coolen, A. M. van den Berg, L. P. Smits, R. Shahzad, D. P. Shamonin, P. J. H. de Koning, A. J. Nederveen, and R. J. van der Geest, "Learning-based automated segmentation of the carotid artery vessel wall in dual-sequence MRI using subdivision surface fitting," Med. Phys., vol. 44, no. 10, pp. 5244-5259, Oct. 2017.
A. M. Arias-Lorza, D. Bos, A. van der Lugt, and M. de Bruijne, "Cooperative carotid artery centerline extraction in MRI," PLoS ONE, vol. 13, no. 5, May 2018, Art. no. e0197180.
R. Zhang, Q. Zhang, A. Ji, P. Lv, J. Zhang, C. Fu, and J. Lin, "Identification of high-risk carotid plaque with MRI-based radiomics and machine learning," Eur. Radiol., vol. 31, no. 5, pp. 3116-3126, May 2021.
J. M. A. Hofman, W. J. Branderhorst, H. M. M. T. Eikelder, V. C. Cappendijk, S. Heeneman, M. E. Kooi, P. A. J. Hilbers, and B. M. T. H. Romeny, "Quantification of atherosclerotic plaque components using in vivo MRI and supervised classifiers," Magn. Reson. Med., vol. 55, no. 4, pp. 790-799, Apr. 2006.
R. van't Klooster, O. Naggara, R. Marsico, J. H. C. Reiber, J.-F. Meder, R. J. van der Geest, E. Touz , and C. Oppenheim, "Automated versus manual in vivo segmentation of carotid plaque MRI," Amer. J. Neuroradiol., vol. 33, no. 8, pp. 1621-1627, Sep. 2012.
F. Liu, D. Xu, M. S. Ferguson, B. Chu, T. Saam, N. Takaya, T. S. Hatsukami, C. Yuan, and W. S. Kerwin, "Automated in vivo segmentation of carotid plaque MRI with morphology-enhanced probability maps," Magn. Reson. Med., Off. J. Int. Soc. Magn. Reson. Med., vol. 55, no. 3, pp. 659-668, 2006.
W. Kerwin, D. Xu, F. Liu, T. Saam, H. Underhill, N. Takaya, B. Chu, T. Hatsukami, and C. Yuan, "Magnetic resonance imaging of carotid atherosclerosis," Topics Magn. Reson. Imag., vol. 18, no. 5, pp. 371-378, 2007.
H. Tang, T. van Walsum, R. S. van Onkelen, R. Hameeteman, S. Klein, M. Schaap, F. L. Tori, Q. J. A. van den Bouwhuijsen, J. C. M. Witteman, A. van der Lugt, L. J. van Vliet, and W. J. Niessen, "Semiautomatic carotid lumen segmentation for quantification of lumen geometry in multispectral MRI," Med. Image Anal., vol. 16, no. 6, pp. 1202-1215, Aug. 2012.
A. van Engelen, W. J. Niessen, S. Klein, H. C. Groen, H. J. M. Verhagen, J. J.Wentzel, A. van der Lugt, and M. de Bruijne, "Atherosclerotic plaque component segmentation in combined carotid MRI and CTA data incorporating class label uncertainty," PLoS ONE, vol. 9, no. 4, Apr. 2014, Art. no. e94840.
H. Tang, M. Selwaness, R. Hameeteman, A. van Dijk, A. van der Lugt, J. C. Witteman, W. J. Niessen, L. J. van Vliet, and T. van Walsum, "Semi-automatic MRI segmentation and volume quantification of intraplaque hemorrhage," Int. J. Comput. Assist. Radiol. Surg., vol. 10, no. 1, pp. 67-74, Jan. 2015.
Y. Dong, Y. Pan, X. Zhao, R. Li, C. Yuan, and W. Xu, "Identifying carotid plaque composition in MRI with convolutional neural networks," in Proc. IEEE Int. Conf. Smart Comput. (SMARTCOMP), May 2017, pp. 1-8.
J. Wu, J. Xin, X. Yang, J. Sun, D. Xu, N. Zheng, and C. Yuan, "Deep morphology aided diagnosis network for segmentation of carotid artery vessel wall and diagnosis of carotid atherosclerosis on blackblood vessel wall MRI," Med. Phys., vol. 46, no. 12, pp. 5544-5561, 2019.
L. Chen, J. Sun, G. Canton, N. Balu, D. S. Hippe, X. Zhao, R. Li, T. S. Hatsukami, J.-N. Hwang, and C. Yuan, "Automated artery localization and vessel wall segmentation using tracklet refinement and polar conversion," IEEE Access, vol. 8, pp. 217603-217614, 2020.
L. Chen, H. Zhao, H. Jiang, N. Balu, D. B. Geleri, B. Chu, H. Watase, X. Zhao, R. Li, J. Xu, T. S. Hatsukami, D. Xu, J.-N. Hwang, and C. Yuan, "Domain adaptive and fully automated carotid artery atherosclerotic lesion detection using an artificial intelligence approach (LATTE) on 3D MRI," Magn. Reson. Med., vol. 86, no. 3, pp. 1662-1673, Sep. 2021.
D. D. Samber, S. Ramachandran, A. Sahota, S. Naidu, A. Pruzan, Z. A. Fayad, and V. Mani, "Segmentation of carotid arterial walls using neural networks," World J. Radiol., vol. 12, no. 1, pp. 1-9, Jan. 2020.
R. Camarasa, D. Bos, J. Hendrikse, P. Nederkoorn, E. Kooi, A. van der Lugt, and M. de Bruijne, "Quantitative comparison of Monte-Carlo dropout uncertainty measures for multi-class segmentation," in Proc. Int. Workshop Uncertainty Safe Utilization Mach. Learn. Med. Imag., 2020, pp. 32-41.
D. Alblas, C. Brune, and J. M. Wolterink, "Deep-learning-based carotid artery vessel wall segmentation in black-blood MRI using anatomical priors," Proc. SPIE, vol. 12032, Apr. 2021, Art. no. 120320Y.
G. Varoquaux and V. Cheplygina, "Machine learning for medical imaging: Methodological failures and recommendations for the future," NPJ Digit. Med., vol. 5, no. 1, p. 48, Apr. 2022.
E. Kondrateva, M. Pominova, E. Popova, M. Sharaev, A. Bernstein, and E. Burnaev, "Domain shift in computer vision models for MRI data analysis: An overview," Proc. SPIE, vol. 11605, Jan. 2020, Art. no. 116050H.
R. T. Shinohara, E. M. Sweeney, J. Goldsmith, N. Shiee, F. J. Mateen, P. A. Calabresi, S. Jarso, D. L. Pham, D. S. Reich, and C. M. Crainiceanu, "Statistical normalization techniques for magnetic resonance imaging," NeuroImage, Clin., vol. 6, pp. 9-19, Jan. 2014.
M. Abdar, F. Pourpanah, S. Hussain, D. Rezazadegan, L. Liu, M. Ghavamzadeh, P. Fieguth, X. Cao, A. Khosravi, U. R. Acharya, V. Makarenkov, and S. Nahavandi, "A review of uncertainty quantification in deep learning: Techniques, applications and challenges," Inf. Fusion, vol. 76, pp. 243-297, Dec. 2021.
C. Zhu, X. Wang, S. Chen, Z. Teng, C. Bai, X. Huang, M. Xia, Z. Shao, Z. Gu, and P. Sun, "Complex carotid artery segmentation in multi-contrast MR sequences by improved optimal surface graph cuts based on flow line learning," Med. Biol. Eng. Comput., vol. 60, no. 9, pp. 2693-2706, Sep. 2022.
F. Isensee, P. F. Jaeger, S. A. A. Kohl, J. Petersen, and K. H. Maier-Hein, "nnU-Net: A self-configuring method for deep learning-based biomedical image segmentation," Nature Methods, vol. 18, no. 2, pp. 203-211, Feb. 2021.
S. Hu, Z. Liao, and Y. Xia, "Label propagation for 3D carotid vessel wall segmentation and atherosclerosis diagnosis," Aug. 2022, arXiv:2208.13337.
L. Lavrova, "Lavrovaliz/ur-cara-net: 0.1," Zenodo, Version 0.1, Mar. 2023, doi: 10.5281/zenodo.7741155.
O. Ronneberger, P. Fischer, and T. Brox, "U-Net: Convolutional networks for biomedical image segmentation," in Proc. Int. Conf. Med. Image Comput. Comput.-Assist. Intervent., 2015, pp. 234-241.
T. W. Arega, S. Bricq, F. Meriaudeau, "Leveraging uncertainty estimates to improve segmentation performance in cardiac MR," in Uncertainty for Safe Utilization of Machine Learning in Medical Imaging, and Perinatal Imaging, Placental and Preterm Image Analysis, C. H. Sudre, R. Licandro, C. Baumgartner, A. Melbourne, A. Dalca, J. Hutter, R. Tanno, E. A. Turk, K. Van Leemput, B. T. Jordina, W. M. Wells, and C. Macgowan, Eds. Cham, Switzerland: Springer, 2021, pp. 24-33.
S. Shit, J. C. Paetzold, A. Sekuboyina, I. Ezhov, A. Unger, A. Zhylka, J. P. W. Pluim, U. Bauer, and B. H. Menze, "clDice-A novel topologypreserving loss function for tubular structure segmentation," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), Jun. 2021, pp. 16560-16569.
A. Reinke et al., "Common limitations of image processing metrics: A picture story," Apr. 2021, arXiv:2104.05642.
F. Vaassen, C. Hazelaar, A. Vaniqui, M. Gooding, B. van der Heyden, R. Canters, andW. van Elmpt, "Evaluation of measures for assessing timesaving of automatic organ-at-risk segmentation in radiotherapy," Phys. Imag. Radiat. Oncol., vol. 13, pp. 1-6, Jan. 2020.
P. F lt, J. Hyttinen, L. Fauch, A. Riepponen, A. Kullaa, and M. Hauta-Kasari, "Spectral image enhancement for the visualization of dental lesions," in Proc. 8th Int. Conf. Image Signal Process., 2018, pp. 490-498.
