Fully automated algorithm for the detection of bone marrow oedema lesions in patients with axial spondyloarthritis – Feasibility study

• Autorzy: Rzecki Krzysztof , Kucybała Iwona , Gut Daniel , Jarosz Aldona , Nabagło Tomasz , Tabor Zbisław , Wojciechowski Wadim

Identyfikatory

DOI

10.1016/j.bbe.2021.05.005

• Języki publikacji: EN

Abstrakty

EN

In the present paper we propose an automated procedure for the detection of bone marrow oedema lesions in patients with axial spondyloarthritis (axSpA). The procedure is based on MRI examination of the sacroiliac joints of 30 patients with confirmed sacroiliitis in the course of axSpA (18 of patients were male, while 12 of patients were female; the mean age of patients was 28.8 ± 9.0 years, range from 18 to 45 years). The segmentation of the sacral and iliac bones is performed using U-Net-like architecture. The subchondral bone regions are found, where inflammatory changes are expected to appear. Convolutional classification architectures are trained to classify image voxels as either being within normal or inflammatory-changed areas. The deep learning-based classification of voxels is compared to a method based on statistical testing. The Dice coefficient for segmentation of subchondral bone was 0.84 (standard deviation 0.06). The sensitivity of the detection of inflammatory changes was 0.88. The specificity of the detection of inflammatory changes was 0.91. The discrepancy between sensitivity and specificity values achieved by the automated method and the human readers is attributed to ‘‘a satisfaction of search” phenomenon. After verification of the automated detections by human readers sensitivity and specificity increased to 0.95 and 0.96, respectively. The Spearman’s correlation coefficient between the volumes of lesions calculated manually and automatically is equal to 0.866 while the intraclass coefficient of correlation ICC(1,1) is equal to 0.947. The study demonstrates that an automated detection of inflammatory lesions with high precision of lesion volume assessment is feasible.

Słowa kluczowe

EN

deep learning; MRI; magnetic resonance; axial spondyloarthritis

PL

uczenie głębokie; MRI; rezonans magnetyczny; spondyloartropatia osiowa

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2021
• Tom: Vol. 41, no. 2
• Strony: 833--853
• Opis fizyczny: Bibliogr. 50 poz., rys., tab., wykr.

Twórcy

autor

Rzecki Krzysztof

• AGH University of Science and Technology, al. Adama Mickiewicza 30, 30-059 Krakow, Poland

autor

Kucybała Iwona

• Department of Radiology, Jagiellonian University Medical College, Krakow, Poland

autor

Gut Daniel

• AGH University of Science and Technology, Krakow, Poland

autor

Jarosz Aldona

• Department of Radiology, St. Ludwik Children’s Hospital, Krakow, Poland

autor

Nabagło Tomasz

• AGH University of Science and Technology, Krakow, Poland

autor

Tabor Zbisław

• AGH University of Science and Technology, Krakow, Poland

autor

Wojciechowski Wadim

• Department of Radiology, Jagiellonian University Medical College, Krakow, Poland

Bibliografia

• [1] Rudwaleit M, van der Heijde D, Landewé R, Listing J, Akkoc N, Brandt J, et al. The development of Assessment of SpondyloArthritis international Society classification criteria for axial spondyloarthritis (part II): validation and final selection. J Ann Rheum Dis. 2009;68:777-783.
• [2] Lukas C, Cyteval C, Dougados M, Weber U. MRI for diagnosis of axial spondyloarthritis: major advance with critical limitations ’Not everything that glisters is gold (standard)’. RMD Open 2018;4(1):e000586.
• [3] MaksymowychWP, Lambert RGW, Østergaard M, Pedersen SJ, Machado PM, Weber U, et al. MRI lesions in the sacroiliac joints of patients with spondyloarthritis: an update of definitions and validation by the ASAS MRI working group. Ann Rheum Dis 2019;78(11):1550–8.
• [4] Akhavan S, Martinkovich SC, Kasik C, DeMeo PJ. Bone marrow edema, clinical significance, and treatment options: a review. J Am Acad Orthop Surg 2020;28(20):e888–99.
• [5] Baumbach SF, Pfahler V, Bechtold-Dalla Pozza S, Feist-Pagenstert I, Fürmetz J, Baur-Melnyk A, et al. How we manage bone marrow edema-an interdisciplinary approach. J Clin Med 2020;9(2):551.
• [6] Nygaard A, Jurik AG, Lund C, Schiøttz-Christensen B. The incidence of bone marrow oedema at the sacroiliac joints in a non-rheumatological population — a retrospective cohort study. BMC Musculoskelet Disord 2019;20:590.
• [7] Bhutani M, Turkbey B, Tan E, Korde N, Kwok M, Manasanch EE, et al. Bone marrow abnormalities and early bone lesions in multiple myeloma and its precursor disease: a prospective study using functional and morphologic imaging. Leukemia Lymphoma 2016;57(5):1114–21.
• [8] Minaie A, Woolley K, Smith R, Ezomo O, Feinn R, Mele B, et al. Detecting bone marrow edema with magnetic resonance spectroscopy: a brief report. J Nurse Practitioners 2020;16(8): e129–32.
• [9] Compagnoni R, Lesman J, Ferrua P, Menon A, Minoli C, Gallazzi M, et al. Validation of a new topographic classification of bone marrow lesions in the knee: the sixletter system. Knee Surg Sports Traumatol Arthrosc 2021;29 (2):333–41.
• [10] Baraliakos X, Richter A, Feldmann D, Ott A, Buelow R, Schmidt CO, et al. Which factors are associated with bone marrow oedema suspicious of axial spondyloarthritis as detected by MRI in the sacroiliac joints and the spine in the general population? Ann Rheum Dis 2021;80:469–74.
• [11] Renson T, Carron P, De Craemer A-S, Deroo L, de Hooge M, Krabbe S, et al. Axial involvement in patients with early peripheral spondyloarthritis: a prospective MRI study of sacroiliac joints and spine. Ann Rheum Dis 2021;80(1):103–8.
• [12] Dube B, Bowes MA, Hensor EMA, Barr A, Kingsbury SR, Conaghan PG. The relationship between two different measures of osteoarthritis bone pathology, bone marrow lesions and 3D bone shape: data from the Osteoarthritis Initiative. Osteoarthritis Cartilage 2018;26(10):1333–7.
• [13] Wang F, Luo A, Xuan W, Qi L, Wu Q, Gan K, et al. The bone marrow edema links to an osteoclastic environment and precedes synovitis during the development of collagen induced arthritis. Front Immunol 2019;10:884.
• [14] Timmer VCML, Kroonenburgh AMJLV, Henneman WJP, Vaassen LAA, Roele ED, Kessler PAWH, et al. Detection of bone marrow edema in the head and neck with dual-energy CT: ready for clinical use? AJR Am J Roentgenol 2020;214 (4):893–9.
• [15] Dalto VF, Assad RL, Crema MD, Louzada-Junior P, Nogueira-Barbosa MH. MRI assessment of bone marrow oedema in the sacroiliac joints of patients with spondyloarthritis: is the SPAIR T2w technique comparable to STIR? Eur Radiol 2017;27(9):3669–76.
• [16] Baraliakos X, Braun J. Imaging scoring methods in axial spondyloarthritis. Rheum Dis Clin North Am 2016;42(4):663–78.
• [17] Sieper J, Rudwaleit M, Baraliakos X, Brandt J, Braun J, Burgos-Vargas R, et al. The Assessment of SpondyloArthritis international Society (ASAS) handbook: a guide to assess spondyloarthritis. Ann Rheum Dis. 2009;68(Suppl 2):ii1-ii44.
• [18] Mandl P, Navarro-Compán V, Terslev L, Aegerter P, van der Heijde D, D’Agostino MA, et al. EULAR recommendations for the use of imaging in the diagnosis and management of spondyloarthritis in clinical practice. Ann Rheum Dis 2015;74 (7):1327–39.
• [19] Landewé RBM, Hermann KGA, Van Der Heijde DMFM, Baraliakos X, Jurik AG, Lambert RG, et al. Scoring sacroiliac joints by magnetic resonance imaging. A multiple-reader reliability experiment. J Rheumatol 2005;32(10):2050–5.
• [20] Van Der Heijde D, Landewé R, Hermann KG, Rudwaleit M, Østergaard M, Oostveen A, et al. Is there a preferred method for scoring activity of the spine by magnetic resonance imaging in ankylosing spondylitis? J Rheumatol 2007;34(4):871–3.
• [21] Maksymowych WP, Inman RD, Salonen D, Dhillon SS, Williams M, Stone M, et al. Spondyloarthritis Research Consortium of Canada magnetic resonance imaging index for assessment of sacroiliac joint inflammation in ankylosing spondylitis. Arthritis Rheum 2005;53(5):703–9.
• [22] Rudwaleit M, Schwarzlose S, Hilgert ES, Listing J, Braun J, Sieper J. MRI in predicting a major clinical response to antitumour necrosis factor treatment in ankylosing spondylitis. Ann Rheum Dis 2008;67(9):1276–81.
• [23] McHugh ML. Interrater reliability: the kappa statistic. Biochem Med (Zagreb) 2012:276–82.
• [24] Hosny A, Parmar C, Quackenbush J, Schwartz LH, Aerts HJWL. Artificial intelligence in radiology. Nat Rev Cancer 2018;18(8):500–10.
• [25] Nishida Y, Kimura S, Mizobe H, Yamamichi J, Kojima K, Kawaguchi A, et al. Automatic digital quantification of bone marrow myeloma volume in appendicular skeletons - clinical implications and prognostic significance. Sci Rep 2017;7(1). https://doi.org/10.1038/s41598-017-13255-w.
• [26] Yu TC, Chou WC, Yeh CY, Yang CK, Huang SC, Tien FM, et al. Automatic bone marrow cell identification and classification by deep neural network. Blood 2019;134 (Supplement_1):2084.
• [27] Anilkumar KK, Manoj VJ, Sagi TM. A survey on image segmentation of blood and bone marrow smear images with emphasis to automated detection of Leukemia. Biocybernetics Biomed Eng 2020;40(4):1406–20.
• [28] Hegde RB, Prasad K, Hebbar H, Sandhya. Peripheral blood smear analysis using image processing approach for diagnostic purposes: a review. Biocybernetics Biomed Eng 2018;38(3):467–80.
• [29] Omiotek Z, Stepanchenko O,WójcikW, Legieć W, Szatkowska M. The use of the Hellwig’s method for feature selection in the detection of myeloma bone destruction based on radiographic images. Biocybernetics Biomed Eng 2019;39 (2):328–38.
• [30] Aprovitola A, Gallo L. Knee bone segmentation from MRI: A classification and literature review. Biocybernetics Biomed Eng 2016;36(2):437–49.
• [31] Saygılı A, Albayrak S. A new computer-based approach for fully automated segmentation of knee meniscus from magnetic resonance images. Biocybernetics Biomed Eng 2017;37(3):432–42.
• [32] Zarco P, Almodóvar R, Bueno Á , Molinero LM. Development and validation of SCAISS, a tool for semi-automated quantification of sacroilitis by magnetic resonance in spondyloarthritis. Rheumatol Int 2018;38(10):1919–26.
• [33] Kucybała I, Tabor Z, Polak J, Urbanik A, Wojciechowski W. The semi-automated algorithm for the detection of bone marrow oedema lesions in patients with axial spondyloarthritis. Rheumatol Int 2020;40:625–33.
• [34] Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al. Fiji: an open-source platform for biological image analysis. Nat Methods 2012;9:676–82.
• [35] Ronneberger O, Fischer P, Brox T. 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. Cham, Switzerland: Springer, 2015;9351:234–241.
• [36] Chen LC, Zhu Y, Papandreou G, Schroff F, Adam H. Encoder-decoder with atrous separable convolution for semantic image segmentation. In: Ferrari V., Hebert M., Sminchisescu C., Weiss Y. (eds) Computer Vision – ECCV 2018. Cham, Switzerland: Springer, Lecture Notes in Computer Science 2018;11211:833-851.
• [37] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. 3rd International Conference on Learning Representations; 2015. San Diego, CA, USA, May 7-9, 2015.
• [38] Yu H, Zhang Z, Qin Z, Wu H, Li D, Zhao J, et al. Loss rank mining: a general hard example mining method for real-time detectors. In: 2018 International Joint Conference on Neural Networks (IJCNN). Rio de Janeiro. p. 1–8.
• [39] He K, Zhang X, Ren S, Sun J. Identity mappings in deep residual networks. In: Leibe B., Matas J, Sebe N, Welling M (eds) Computer Vision – ECCV 2016. Cham, Switzerland: Springer, Lecture Notes in Computer Science 2016;9908:630-645.
• [40] Szegedy C, Wei L, Yangqing J, Sermanet P, Reed S, Anguelov D, et al. Going deeper with convolutions. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston, MA, 2015, pp. 1-9.
• [41] Berbaum KS, Franken EA, Dorfman DD, Rooholamini SA, Kathol MH, Barloon TJ, et al. Satisfaction of search in diagnostic radiology. Invest Radiol 1990;25(2):133–9.
• [42] Weiss PF, Maksymowych WP, Lambert RG, Jaremko JL, Biko DM, Paschke J, et al. Feasibility and reliability of the Spondyloarthritis Research Consortium of Canada sacroiliac joint inflammation score in children. Arthritis Res Ther 2018;20(1). https://doi.org/10.1186/s13075-018-1543-x.
• [43] Carmona L, Sellas A, Rodríguez-Lozano C, Juanola X, Llorente JFG, Sueiro JLF, et al. Scoring with the Berlin MRI method for assessment of spinal inflammatory activity in patients with ankylosing spondylitis: a calibration exercise among rheumatologists. Clin Exp Rheumatol 2013;31(6):883–8.
• [44] Isensee F, Petersen J, Klein A, Zimmerer D, Jaeger PF, Kohl S, et al.: nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation. arXiv:1809.10486v1, 2018.
• [45] Isensee F, Jaeger PF, Kohl S, Petersen J, Maier-Hein KH. Automated Design of Deep Learning Methods for Biomedical Image Segmentation. arXiv:1904.08128v2, 2020.
• [46] Zhou Z, Siddiquee MMR, Tajbakhsh N, Liang J. UNet++: Redesigning skip connections to exploit multiscale features in image segmentation. IEEE Trans Med Imaging 2020;39(6):1856–67.
• [47] Huang H, Lin L, Tong R, Hu H, Zhang Q, Iwamoto Y, et al. UNet 3+: A Full-Scale Connected UNet for Medical Image Segmentation. arXiv:2004.08790v1, 2020.
• [48] Feng S, Zhao H, Shi F, Cheng X, Wang M, Ma Y, et al. CPFNet: Context pyramid fusion network for medical image segmentation. IEEE Trans Med Imaging 2020;39(10):3008–18.
• [49] Mou L, Zhao Y, Fu H, Liu Y, Cheng J, Zheng Y, et al. CS2-Net: Deep learning segmentation of curvilinear structures in medical imaging. Med Image Anal 2021;67 101874.
• [50] Selvaraju RR, Das A, Vedantam R, Cogswell M, Parikh D, Batra D. Grad-CAM: Why did you say that? Visual Explanations from Deep Networks via Gradient-based Localization. arXiv:1610.02391v4, 2016.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2021).