Journal of Korean Neurosurgical Society

The Korean Neurosurgical Society (대한신경외과학회)
권호 표지
DOI QR코드

DOI QR Code

Automatically Diagnosing Skull Fractures Using an Object Detection Method and Deep Learning Algorithm in Plain Radiography Images

  • Tae Seok, Jeong (Department of Traumatology, Gil Medical Center, Gachon University College of Medicine) ;
  • Gi Taek, Yee (Department of Neurosurgery, Gil Medical Center, Gachon University College of Medicine) ;
  • Kwang Gi, Kim (Department of Biomedical Engineering, Gil Medical Center, Gachon University College of Medicine) ;
  • Young Jae, Kim (Department of Biomedical Engineering, Gil Medical Center, Gachon University College of Medicine) ;
  • Sang Gu, Lee (Department of Neurosurgery, Gil Medical Center, Gachon University College of Medicine) ;
  • Woo Kyung, Kim (Department of Traumatology, Gil Medical Center, Gachon University College of Medicine)
  • Received : 2022.03.22
  • Accepted : 2022.05.18
  • Published : 2023.01.01
Download PDF
⟨ Previous Next ⟩

Abstract

Objective : Deep learning is a machine learning approach based on artificial neural network training, and object detection algorithm using deep learning is used as the most powerful tool in image analysis. We analyzed and evaluated the diagnostic performance of a deep learning algorithm to identify skull fractures in plain radiographic images and investigated its clinical applicability. Methods : A total of 2026 plain radiographic images of the skull (fracture, 991; normal, 1035) were obtained from 741 patients. The RetinaNet architecture was used as a deep learning model. Precision, recall, and average precision were measured to evaluate the deep learning algorithm's diagnostic performance. Results : In ResNet-152, the average precision for intersection over union (IOU) 0.1, 0.3, and 0.5, were 0.7240, 0.6698, and 0.3687, respectively. When the intersection over union (IOU) and confidence threshold were 0.1, the precision was 0.7292, and the recall was 0.7650. When the IOU threshold was 0.1, and the confidence threshold was 0.6, the true and false rates were 82.9% and 17.1%, respectively. There were significant differences in the true/false and false-positive/false-negative ratios between the anterior-posterior, towne, and both lateral views (p=0.032 and p=0.003). Objects detected in false positives had vascular grooves and suture lines. In false negatives, the detection performance of the diastatic fractures, fractures crossing the suture line, and fractures around the vascular grooves and orbit was poor. Conclusion : The object detection algorithm applied with deep learning is expected to be a valuable tool in diagnosing skull fractures.

Keywords

Acknowledgement

This work was supported by Institute of Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) (No. 2020-0-00161-001, Active Machine Learning based on Open-set training for Surgical Video) and was supported by the Gachon Gil Medical Center (FRD2019-11-02(3)) and GRRC program of Gyeonggi province (No. GRRC-Gachon2020(B01)).

References

  1. Amyot F, Arciniegas DB, Brazaitis MP, Curley KC, Diaz-Arrastia R, Gandjbakhche A, et al. : A review of the effectiveness of neuroimaging modalities for the detection of traumatic brain injury. J Neurotrauma 32 : 1693-1721, 2015 https://doi.org/10.1089/neu.2013.3306
  2. Beyaz S, Acici K, Sumer E : Femoral neck fracture detection in X-ray images using deep learning and genetic algorithm approaches. Jt Dis Relat Surg 31 : 175-183, 2020 https://doi.org/10.5606/ehc.2020.72163
  3. Bruns JJ Jr, Jagoda AS : Mild traumatic brain injury. Mt Sinai J Med 76 : 129-137, 2009 https://doi.org/10.1002/msj.20101
  4. Chung SW, Han SS, Lee JW, Oh KS, Kim NR, Yoon JP, et al. : Automated detection and classification of the proximal humerus fracture by using deep learning algorithm. Acta orthop 89 : 468-473, 2018 https://doi.org/10.1080/17453674.2018.1453714
  5. Everingham M, Van Gool L, Williams CK, Winn J, Zisserman A : The pascal visual object classes (voc) challenge. Int J Comput Vis 88 : 303-338, 2010 https://doi.org/10.1007/s11263-009-0275-4
  6. He K, Zhang X, Ren S, Sun J : Deep residual learning for image recognition : Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway : The Institute of Electrical and Electronics Engineers, Inc., 2016, pp770-778
  7. Kim HJ, Roh HG, Lee IW : Craniosynostosis : updates in radiologic diagnosis. J Korean Neurosurg Soc 59 : 219-226, 2016 https://doi.org/10.3340/jkns.2016.59.3.219
  8. Krogue JD, Cheng KV, Hwang KM, Toogood P, Meinberg EG, Geiger EJ, et al. : Automatic hip fracture identification and functional subclassification with deep learning. Radiol Artif Intell 2 : e190023, 2020 https://doi.org/10.1148/ryai.2020190023
  9. Lakhani P, Sundaram B : Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks. Radiology 284 : 574-582, 2017 https://doi.org/10.1148/radiol.2017162326
  10. Le TH, Gean AD : Neuroimaging of traumatic brain injury. Mt Sinai J Med 76 : 145-162, 2009 https://doi.org/10.1002/msj.20102
  11. Lin TY, Goyal P, Girshick R, He K, Dollar P : Focal loss for dense object detection : Proceedings of the IEEE International Conference on Computer Vision. Piscataway : The Institute of Electrical and Electronics Engineers, Inc., 2017, pp2980-2988
  12. Lin TY, Maire M, Belongie S, Hays J, Perona P, Ramanan D, et al. : Microsoft coco: common objects in context : European Conference on Computer Vision. Cham : Springer, 2014, pp740-755
  13. Moller TB, Reif E : Pocket Atlas of Radiographic Positioning: Including Positioning for Conventional Angiography, CT, and MRI. Noida : Thieme, 2008
  14. Nakahara K, Utsuki S, Shimizu S, Iida H, Miyasaka Y, Takagi H, et al. : Age dependence of fusion of primary occipital sutures: a radiographic study. Childs Nerv Syst 22 : 1457-1459, 2006 https://doi.org/10.1007/s00381-006-0210-8
  15. Ozkaya E, Topal FE, Bulut T, Gursoy M, Ozuysal M, Karakaya Z : Evaluation of an artificial intelligence system for diagnosing scaphoid fracture on direct radiography. Eur J Trauma Emerg Surg 48 : 585-592, 2020 https://doi.org/10.1007/s00068-020-01468-0
  16. Rezatofighi H, Tsoi N, Gwak J, Sadeghian A, Reid I, Savarese S : Generalized intersection over union: a metric and a loss for bounding box regression : Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway : The Institute of Electrical and Electronics Engineers, Inc., 2019, pp658-666
  17. Sanchez T, Stewart D, Walvick M, Swischuk L : Skull fracture vs. accessory sutures: how can we tell the difference? Emerg Radiol 17 : 413-418, 2010 https://doi.org/10.1007/s10140-010-0877-8
  18. Sim SY, Yoon SH, Kim SY : Quantitative analysis of developmental process of cranial suture in Korean infants. J Korean Neurosurg Soc 51 : 31-36, 2012 https://doi.org/10.3340/jkns.2012.51.1.31
  19. Taylor CA, Bell JM, Breiding MJ, Xu L : Traumatic brain injury-related emergency department visits, hospitalizations, and deaths - united states, 2007 and 2013. MMWR Surveill Summ 66 : 1-16, 2017 https://doi.org/10.15585/mmwr.ss6609a1
  20. Weir P, Suttner NJ, Flynn P, McAuley D : Normal skull suture variant mimicking intentional injury. BMJ 332 : 1020-1021, 2006 https://doi.org/10.1136/bmj.332.7548.1020
  21. Wintermark M, Sanelli PC, Anzai Y, Tsiouris AJ, Whitlow CT; ACR Head Injury Institute, et al. : Imaging evidence and recommendations for traumatic brain injury: conventional neuroimaging techniques. J Am Coll Radiol 12 : e1-e14, 2015
  22. Zou Z, Shi Z, Guo Y, Ye J : Object detection in 20 years: a survey. Available at : https://doi.org/10.48550/arXiv.1905.05055