Artificial intelligence for diagnosis of vertebral compression fractures using a morphometric analysis model, based on convolutional neural networks

BACKGROUND: Pathological low-energy (LE) vertebral compression fractures (VFs) are common complications of osteoporosis and predictors of subsequent LE fractures. In 84% of cases, VFs are not reported on chest CT (CCT), which calls for the development of an artificial intelligence-based (AI) assistant that would help radiology specialists to improve the diagnosis of osteoporosis complications and prevent new LE fractures.
AIMS: To develop an AI model for automated diagnosis of compression fractures of the thoracic spine based on chest CT images.
MATERIALS AND METHODS: Between September 2019 and May 2020 the authors performed a retrospective sampling study of ССТ images. The 160 of results were selected and anonymized. The data was labeled by seven readers. Using the morphometric analysis, the investigators received the following metric data: ventral, medial and dorsal dimensions. This was followed by a semiquantitative assessment of VFs degree. The data was used to develop the Comprise-G AI mode based on CNN, which subsequently measured the size of the vertebral bodies and then calculates the compression degree. The model was evaluated with the ROC curve analysis and by calculating sensitivity and specificity values.
RESULTS: Formed data consist of 160 patients (a training group - 100 patients; a test group - 60 patients). The total of 2,066 vertebrae was annotated. When detecting Grade 2 and 3 maximum VFs in patients the Comprise-G model demonstrated sensitivity - 90,7%, specificity - 90,7%, AUC ROC - 0.974 on the 5-FOLD cross-validation data of the training dataset; on the test data - sensitivity - 83,2%, specificity - 90,0%, AUC ROC - 0.956; in vertebrae demonstrated sensitivity - 91,5%, specificity - 95,2%, AUC ROC - 0.981 on the cross-validation data; for the test data sensitivity - 79,3%, specificity - 98,7%, AUC ROC - 0.978.
CONCLUSIONS: The Comprise-G model demonstrated high diagnostic capabilities in detecting the VFs on CCT images and can be recommended for further validation.

1. Mel'nichenko GA, Belaya ZhE, Rozhinskaya LYa, et al. Russian federal clinical guidelines on the diagnostics, treatment, and prevention of osteoporosis. Problemy endocrinologii. 2017;63(6):392-426. (in Russ.). doi: 10.14341/probl2017636392-426

2. Lesnyak OM, Baranova IA, Belova KYu et al. Osteoporosis in Russian Federation: Epidemiology, Socio-Medical and Economical Aspects (Review). Travmatologiya i ortopediya Rossii [Traumatology and orthopedics of Russia]. 2018;24(1):155-168. (in Russ.). doi: 10.21823/2311-2905-2018-24-1-155-168

3. Lesnyak OM. The Eastern European & Central Asian Regional Audit Epidemiology, costs & burden of osteoporosis in 2010. Osteoporosis and bone diseases. 2011;14(2):3-6. (in Russ.). doi: 10.14341/osteo201123-6

4. Kanis JA, McCloskey EV, Johansson H, et al. European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporosis International. 2013; 24:23-57. doi: 10.1007/s00198-012-2074-y

5. Williams AL, Al-Busaidi A, Sparrow PJ, et al. Underreporting of osteoporotic vertebral fractures on computed tomography. European journal of radiology. 2009;69(1):179-183. doi: 10.1016/j.ejrad.2007.08.028

6. Carberry GA, Pooler BD, Binkley N, et al. Unreported vertebral body compression fractures at abdominal multidetector CT. Radiology. 2013;268(1):120-126. doi: 10.1148/radiol.13121632

7. Black DM, Arden NK, Palermo L, et al. Prevalent vertebral deformities predict hip fractures and new vertebral deformities but not wrist fractures. J. Bone Miner. Res. 1999;14(5): 821-828. doi: 10.1359/jbmr.1999.14.5.821

8. Ahmed N, Mandel R, Fain MJ. Frailty: An Emerging Geriatric Syndrome. The American Journal of Medicine. 2007;120(9):748-753. doi: 10.1016/j.amjmed.2006.10.018

9. Antonio CB, Bautista LGC, Labao AB et al. Vertebra fracture classification from 3D CT lumbar spine segmentation masks using a convolutional neural network. 10th Asian Conference on Intelligent Information and Database Systems, ACIIDS 2018. 2018 March 19-21. Dong Hoi City, Vietnam;449-458. doi: 10.1007/978-3-319-75420-8_43

10. Valentinitsch A, Trebeschi S, Kaesmacher J, et al. Opportunistic osteoporosis screening in multi-detector ct images via local classification of textures. Osteoporosis international. 2019;30(6):1275-1285. doi: 10.1007/s00198-019-04910-1

11. Nicolaes J, Raeymaeckers S, Robben D, et al. Detection of vertebral fractures in CT using 3D Convolutional Neural Networks. Computational Methods and Clinical Applications for Spine Imaging. 2019:3-14. doi: 10.1007/978-3-030-39752-4_1

12. Burns JE, Yao J, Summers RM. Vertebral body compression fractures and bone density: automated detection and classification on CT images. Radiology. 2017;284(3):788-797. doi: 10.1148/radiol.2017162100

13. Roth HR, Wang Y, Yao J, et al. Deep convolutional networks for automated detection of posterior-element fractures on spine CT. Medical Imaging 2016: Computer-Aided Diagnosis. 2016. doi: 10.1117/12.2217146

14. Bar A, Wolf L, Amitai OB, et al. Compression fractures detection on CT. Medical Imaging 2017: Computer-Aided Diagnosis. 2017. doi:10.1117/12.2249635

15. Tomita N, Cheung YY, Hassanpour S. Deep neural networks for automatic detection of osteoporotic vertebral fractures on CT scans. Computers in biology and medicine. 2018;(98);8-15. doi: 10.1016/j.compbiomed.2018.05.011

16. Genant HK, Wu CY, Van KC, et al. Vertebral fracture assessment using a semiquantitative technique. Journal of bone and mineral research. 1993;8(9):1137-1148. doi: 10.1002/jbmr.5650080915

17. Decree of the President of Russian Federation N490 of 10 October 2019. O razvitii iskusstvennogo intellekta v Rossiiskoi Federatsii». (In Russ.). Доступно по: http://www.kremlin.ru/acts/bank/44731. Ссылка активна на 03.09.2020.

18. Adams JE, Lenchik L, Roux C, Genant HK. Vertebral Fracture Initiative. Part II. Radiological Assessment of Vertebral Fracture. [Internet]. [cited 2020 Aug 5]; 49 p. Available from: https://www.iofbonehealth.org/sites/default/files/PDFs/Vertebral%20Fracture%20Initiative/IOF_VFI-Part_II-Manuscript.pdf

19. Alqahtani FF, Offiah AC. Diagnosis of osteoporotic vertebral fractures in children. Pediatric Radiology. 2019;49(3);283-296. doi: 10.1007/s00247-018-4279-5

20. Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. Medical Image Computing and Computer-Assisted Intervention — MICCAI 2015. 2015 October 4-9. Munich, Germany; 234-241. doi: 10.1007/978-3-319-24574-4_28

21. Kingma DP, Ba JL. Adam: A method for stochastic optimization. 3rd International Conference on Learning Representations, ICLR 2015 — Conference Track Proceedings [Internet]. 2015 May 7-9. San Diego, USA. Available from: https://arxiv.org/pdf/1412.6980.pdf

22. Redmon J, Farhadi A. Yolov3: An incremental improvement. [Internet]. [cited 2020 Jul 20]; 6 p. Available from: https://arxiv.org/pdf/1804.02767.pdf

23. Pisov M, Kondratenko V, Zakharov A et al. Keypoints localization for joint vertebra detection and fracture severity quantification. [Internet]. [cited 2020 Jul 20]; 13 p. Available from: https://arxiv.org/pdf/2005.11960.pdf

24. Derkatch S, Kirby C, Kimelman D, Jozani MJ, et al. Identification of Vertebral Fractures by Convolutional Neural Networks to Predict Nonvertebral and Hip Fractures: A Registry-based Cohort Study of Dual X-ray Absorptiometry. Radiology. 2019;293(2):190-201. doi: 10.1148/radiol.2019190201

25. Kim DH, MacKinnon T. Artificial intelligence in fracture detection: transfer learning from deep convolutional neural networks. Clinical Radiology. 2018;73(5):439-445. doi: 10.1016/j.crad.2017.11.015

26. Pickhardt PJ, Pooler BD, Lauder T, et al. Opportunistic Screening for Osteoporosis Using Abdominal Computed Tomography Scans Obtained for Other Indications. Annals of Internal Medicine. 2013;158(8):588-595. doi: 10.7326/0003-4819-158-8-201304160-00003.

27. Alacreu E, Moratal D, Arana E. Opportunistic screening for osteoporosis by routine CT in Southern Europe. Osteoporosis International. 2017;28(3):983-990. doi: 10.1007/s00198-016-3804-3