Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (2024)

J Appl Biomed 18:97-105, 2020|DOI:10.32725/jab.2020.013

Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method

Sun-Jung Yoon^{1 #}, Tae Hyong Kim^{2 #}, Su-Bin Joo³, Seung Eel Oh^{4 *}: ¹Jeonbuk National University, Research Institute of Clinical Medicine, Department of Orthopedic Surgery, Jeonju, South Korea; ²Sungkyunkwan University, College of Biotechnology and Bioengineering, Suwon, South Korea; ³Korea Institute of Machinery & Materials, Daegu Research Center for Medical Devices and Green Energy, Daegu, South Korea; ⁴Korea Food Research Institute, Research Group of Consumer Safety, Wanju, South Korea

Intertrochanteric (IT) femur fractures are the most common fractures in elderly people, and they lead to significant morbidity, mortality, and reduced quality of life. The different types of fractures require a careful definition to ensure accurate surgical planning and reduce the operation time, healing time, and number of surgical failures. In this study, a deep learning-based automatic multi-class IT fracture detection model was developed using computed tomography (CT) images and based on the AO/OTA classification method. The original CT image was resized and rearranged according to the fracture location and an unsharp masking filter was applied. A multi-class classification of nine different types of IT fractures and no fracture was performed using the faster regional-convolutional neural network (R-CNN). Bayesian optimization was also implemented to determine the optimal hyperparameter values for the faster R-CNN algorithm. In our proposed model, IT fractures classified into two classes showed an average accuracy of 0.97 ± 0.02, which was 0.90 ± 0.02 when classified into ten classes. Additionally, the detected region of interest from our proposed model showed minimum root mean square error and intersection over union values of 16.34 ± 47.01 pixels and 0.87 ± 0.12, respectively. In the future, our proposed automatic multi-class IT femur fracture detection model could allow clinicians to identify the fracture region and diagnose different types of femur fractures faster and more accurately. This will increase the probability of correct surgical treatment and minimize postoperative complications.

Keywords: AO, OTA classification method; Computer-aided Diagnostic Detection; Deep learning; Intertrochanteric femur fracture; Optimization

Grants and funding:

This research was supported with funding from the Biomedical Research Institute, Chonbuk National University Hospital and Main Research Program (E0162500) of the Korea Food Yoon et al. / J Appl Biomed (A) Expected ROI Detected ROI (B) Expected ROI Detected ROI (C) Expected ROI Detected ROI 105 Research Institute (KFRI), funded by the Ministry of Science and ICT.

Conflicts of interest:

The authors have no conflict of interests to declare.

Received: August 19, 2019; Revised: March 23, 2020; Accepted: August 26, 2020; Prepublished online: September 22, 2020; Published: December 14, 2020 Show citation

ACS

AIP

APA

ASA

Harvard

Chicago

IEEE

ISO690

MLA

NLM

Turabian

Vancouver

Yoon S, Hyong Kim T, Joo S, Eel Oh S. Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method. J Appl Biomed. 2020;18(4):97-105. doi:10.32725/jab.2020.013. PubMed PMID:34907762.

References

Bayram F, Çakiroğlu M (2016). Diffract: Diaphyseal femur fracture classifier system. Biocybern Biomed Eng 36(1): 157-171. DOI: 10.1016/j.bbe.2015.10.003. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (2)$ Go to original source...
Boone C, Carlberg KN, Koueiter DM, Baker KC, Sadowski J, Wiater PJ, et al. (2014). Short versus long intramedullary nails for treatment of intertrochanteric femur fractures (OTA 31-A1 and A2). J Orthop Trauma 28(5): 96-100. DOI: 10.1097/BOT.0b013e3182a7131c. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (3)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (4)$ Go to PubMed...
Braun BJ, Holstein JH, Pohlemann T (2018). Intertrochanteric Hip Fracture: Intramedullary Nails. In: Egol KA, Leucht P (Eds), Proximal Femur Fractures. Springer, Cham, pp. 85-100. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (5)$ Go to original source...
Cho YC, Lee PY, Lee CH, Chen CH, Lin YM (2018). Three-dimensional CT Improves the Reproducibility of Stability Evaluation for Intertrochanteric Fractures. Orthop Surg 10(3): 212-217. DOI: 10.1111/os.12396. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (6)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (7)$ Go to PubMed...
Crijns TJ, Janssen SJ, Davis JT, Ring D, Sanchez HB, Althausen P, et al. (2018). Reliability of the classification of proximal femur fractures: Does clinical experience matter? Injury 49(4): 819-823. DOI: 10.1016/j.injury.2018.02.023. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (8)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (9)$ Go to PubMed...
Erickson BJ, Korfiatis P, Akkus Z, Kline TL (2017). Machine learning for medical imaging. Radiographics 37(2): 505-515. DOI: 10.1148/rg.2017160130. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (10)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (11)$ Go to PubMed...
Faisal M, Nistane P (2016). Proximal Femoral Nailing vs. Dynamic Hip Screw in unstable Intertrochanteric Fracture of Femur - A comparative analysis. Int J Biomed Adv Res 7(10): 489-492. DOI: 10.7439/ijbar. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (12)$ Go to original source...
Fung W, Jönsson A, Bühren V, Bhandari M (2007). Classifying intertrochanteric fractures of the proximal femur: does experience matter? Med Princ Practi 16(3): 198-202. DOI: 10.1159/000100390. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (13)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (14)$ Go to PubMed...
Guan Q, Huang Y, Zhong Z, Zheng Z, Zheng L, Yang Y (2020). Thorax disease classification with attention guided convolutional neural network. Pattern Recognition Letters 131: 38-45. DOI: 10.1016/j.patrec.2019.11.040. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (15)$ Go to original source...
Han SK, Lee BY, Kim YS, Choi NY (2010). Usefulness of multi-detector CT in Boyd-Griffin type 2 intertrochanteric fractures with clinical correlation. Skelet Radiol 39(6): 543-549. DOI: 10.1007/s00256-009-0795-6. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (16)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (17)$ Go to PubMed...
Isida R, Bariatinsky V, Kern G, Dereudre G, Demondion X, Chantelot C (2015). Prospective study of the reproducibility of X-rays and CT scans for assessing trochanteric fracture comminution in the elderly: a series of 110 cases. Eur J Orthop Surg Traumatol 25: 1165-1170. DOI: 10.1007/s00590-015- 1666-6. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (18)$ Go to original source...
Jaderberg M, Simonyan K, Zisserman A (2015). Spatial transformer networks. In: Cortes C, Lee DD, Garnett R, Lawrence ND, Sugiyama M (Eds), Advances in neural information processing systems 28. Montreal, pp. 2017-2025.
Jin WJ, Dai LY, Cui YM, Zhou Q, Jiang LS, Lu H (2005). Reliability of classification systems for intertrochanteric fractures of the proximal femur in experienced orthopaedic surgeons. Injury 36(7): 858-861. DOI: 10.1016/j.injury.2005.02.005. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (19)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (20)$ Go to PubMed...
Kazi A, Albarqouni S, Sanchez AJ, Kirchhoff S, Biberthaler P, Navab N, et al. (2017). Automatic classification of proximal femur fractures based on attention models. In: Wang Q, Shi Y, Suk HI, Suzuki K (Eds), Machine Learning in Medical Imaging. MLMI 2017. Lecture Notes in Computer Science, vol. 10541. Springer, Cham. 70-78. DOI: 10.1007/978-3-319-67389-9_9. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (21)$ Go to original source...
Ker J, Wang L, Rao J, Lim T (2017). Deep learning applications in medical image analysis. IEEE Access 6: 9375-9389. DOI: 10.1109/ACCESS.2017.2788044. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (22)$ Go to original source...
Kim DH, MacKinnon T (2018). Artificial intelligence in fracture detection: transfer learning from deep convolutional neural networks. Clin Radiol 73(5): 439-445. DOI: 10.1016/j.crad.2017.11.015. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (23)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (24)$ Go to PubMed...
Mears SC, Kates SL (2015). A guide to improving the care of patients with fragility fractures, edition 2. Geriatr Orthop Surg Rehab 6(2): 58-120. DOI: 10.1177/2151458515572697. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (25)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (26)$ Go to PubMed...
Pranata YD, Wang WC, Wang JC, Idram I, Lai JY, Liu JW, et al. (2019). Deep learning and SURF for automated classification and detection of calcaneus fractures in CT images. Computer Methods and Programs in Biomedicines 171: 27-37. DOI: 10.1016/j.cmpb.2019.02.006. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (27)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (28)$ Go to PubMed...
Redmon J, Divvala S, Girshick R, Farhadi A (2016). You only look once: Unified, real-time object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 779-788. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (29)$ Go to original source...
Ren S, He K, Girshick R, Sun J (2015). Faster R-CNN: Towards real-time object detection with region proposal networks. In: Advances in neural information processing systems, pp. 91-99.
Segal D, Palmonovich E, Faour A, Marom E, Feldman V, Yaacobi E, et al. (2018). Routine early post-operative X-ray following internal fixation of intertrochanteric femoral fractures is unjustified: a quality improvement study. J Orthop Surg Res 13(1):189. DOI: 10.1186/s13018-018-0896-9. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (30)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (31)$ Go to PubMed...
Shahriari B, Swersky K, Wang Z, Adams RP, de Freitas N (2015). Taking the human out of the loop: A review of Bayesian optimization. Proceedings of the IEEE 104(1): 148-175. DOI: 10.1109/JPROC.2015.2494218. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (32)$ Go to original source...
Shen D, Wu G, Suk HI (2017). Deep learning in medical image analysis. Annu Rev Biomed Eng 19: 221-248. DOI: 10.1146/annurev-bioeng-071516-044442. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (33)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (34)$ Go to PubMed...
Sinno K, Sakr M, Girard J, Khatib H (2010). The effectiveness of primary bipolar arthroplasty in treatment of unstable intertrochanteric fractures in elderly patients. N Am J Med Sci 2(12): 561. DOI: 10.4297/najms.2010.2561. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (35)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (36)$ Go to PubMed...
Toderici G, Vincent D, Johnston N, Jin H, Minnen D, Shor J, et al. (2017). Full resolution image compression with recurrent neural network. In: Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, pp. 5306-5314. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (37)$ Go to original source...
Urakawa T, Tanaka Y, Goto, S, Matsuzawa H, Watanabe K, Endo N (2019). Detecting intertrochanteric hip fracture with orthopedist-level accuracy using a deep convolutional neural network. Skeletal Radiol 48(2): 239-244. DOI: 10.1007/s00256-018-3016-3. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (38)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (39)$ Go to PubMed...
Vedaldi A, Lenc K (2015). Matconvnet: Convolutional neural networks for matlab. In: Proceedings of the 23rd ACM international conference on Multimedia, pp. 689-692. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (40)$ Go to original source...
Voulodimos A, Doulamis N, Doulamis A, Protopapadakis E (2018). Deep learning for computer vision: A brief review. Computational Intelligence and Neuroscience. DOI: 10.1155/2018/7068349. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (41)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (42)$ Go to PubMed...
Wu B, Iandola F, Jin PH, Keutzer K (2017). Squeezedet: Unified, small, low power fully convolutional neural networks for real-time object detection for autonomous driving. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp. 129-137. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (43)$ Go to original source...
Wu J, Davuluri P, Ward KR, co*ckrell C, Hobson R, Najarian K (2012). Fracture detection in traumatic pelvic CT images. Int J Biomed Imaging. DOI: 10.1155/2012/327198. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (44)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (45)$ Go to PubMed...
Yates EJ, Yates LC, Harvey H (2018). Machine learning "red dot": open-source, cloud, deep convolutional neural networks in chest radiograph binary normality classification. Clin Radiol 73(9): 827-831. DOI: 10.1016/j.crad.2018.05.015. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (46)$ Go to original source... $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (47)$ Go to PubMed...
Yu W, Yang K, Bai Y, Xiao T, Yao H, Rui Y (2016). Visualizing and comparing AlexNet and VGG using deconvolutional layers. In: Proceedings of the 33rd International Conference on Machine Learning.
Zhang L, Lin L, Liang X, He K (2016). Is faster R-CNN doing well for pedestrian detection? In: European conference on computer vision October, pp. 443-457. DOI: 10.1007/978-3-319-46475-6_28. $Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method (48)$ Go to original source...

This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0), which permits non-comercial use, distribution, and reproduction in any medium, provided the original publication is properly cited. No use, distribution or reproduction is permitted which does not comply with these terms.

Return to the content