Accuracy of one-step automated orthodontic diagnosis model using a convolutional neural network and lateral cephalogram images with different qualities obtained from nationwide multi-hospitals
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Objective: The purpose of this study was to investigate the accuracy of one-step automated orthodontic diagnosis of skeletodental discrepancies using a convolutional neural network (CNN) and lateral cephalogram images with different qualities from nationwide multi-hospitals. Methods: Among 2,174 lateral cephalograms, 1,993 cephalograms from two hospitals were used for training and internal test sets and 181 cephalograms from eight other hospitals were used for an external test set. They were divided into three classification groups according to anteroposterior skeletal discrepancies (Class I, II, and III), vertical skeletal discrepancies (normodivergent, hypodivergent, and hyperdivergent patterns), and vertical dental discrepancies (normal overbite, deep bite, and open bite) as a gold standard. Pre-trained DenseNet-169 was used as a CNN classifier model. Diagnostic performance was evaluated by receiver operating characteristic (ROC) analysis, t-stochastic neighbor embedding (t-SNE), and gradient-weighted class activation mapping (Grad-CAM). Results: In the ROC analysis, the mean area under the curve and the mean accuracy of all classifications were high with both internal and external test sets (all, > 0.89 and > 0.80). In the t-SNE analysis, our model succeeded in creating good separation between three classification groups. Grad-CAM figures showed differences in the location and size of the focus areas between three classification groups in each diagnosis. Conclusions: Since the accuracy of our model was validated with both internal and external test sets, it shows the possible usefulness of a one-step automated orthodontic diagnosis tool using a CNN model. However, it still needs technical improvement in terms of classifying vertical dental discrepancies.
Abstract
Objective: The purpose of this study was to investigate the accuracy of one-step automated orthodontic diagnosis of skeletodental discrepancies using a convolutional neural network (CNN) and lateral cephalogram images with different qualities from nationwide multi-hospitals. Methods: Among 2,174 lateral cephalograms, 1,993 cephalograms from two hospitals were used for training and internal test sets and 181 cephalograms from eight other hospitals were used for an external test set. They were divided into three classification groups according to anteroposterior skeletal discrepancies (Class I, II, and III), vertical skeletal discrepancies (normodivergent, hypodivergent, and hyperdivergent patterns), and vertical dental discrepancies (normal overbite, deep bite, and open bite) as a gold standard. Pre-trained DenseNet-169 was used as a CNN classifier model. Diagnostic performance was evaluated by receiver operating characteristic (ROC) analysis, t-stochastic neighbor embedding (t-SNE), and gradient-weighted class activation mapping (Grad-CAM). Results: In the ROC analysis, the mean area under the curve and the mean accuracy of all classifications were high with both internal and external test sets (all, > 0.89 and > 0.80). In the t-SNE analysis, our model succeeded in creating good separation between three classification groups. Grad-CAM figures showed differences in the location and size of the focus areas between three classification groups in each diagnosis. Conclusions: Since the accuracy of our model was validated with both internal and external test sets, it shows the possible usefulness of a one-step automated orthodontic diagnosis tool using a CNN model. However, it still needs technical improvement in terms of classifying vertical dental discrepancies.
참고문헌 (25)
- Arık SÖ, "Fully automated quantitative cephalometry using convolutional neural networks", J Med Imaging(Bellingham), 4, pp.014501~ (2017)
- Kim H, "Webbased fully automated cephalometric analysis by deep learning", Comput Methods Programs Biomed, 194, pp.105513~ (2020)
- Nishimoto S, "Personal computer-based cephalometric landmark detection with deep learning, using cephalograms on the internet", J Craniofac Surg, 30, pp.91~95 (2019)
- Erkan M, "Reliability of four different computerized cephalometric analysis programs", Eur J Orthod, 34, pp.318~321 (2012)
- Wen J, "Comparative study of cephalometric measurements using 3imaging modalities", J Am Dent Assoc, 148, pp.913~921 (2017)
- Rudolph DJ, "Automatic computerized radiographic identification of cephalometric landmarks", Am J Orthod Dentofacial Orthop, 113, pp.173~179 (1998)
- Mosleh MA, "Ceph-X: development and evaluation of 2D cephalometric system", BMC Bioinformatics, 17(Suppl 19), pp.499~ (2016)
- Yu HJ, "Automated skeletal classification with lateral cephalometry based on artificial intelligence", J Dent Res, 99, pp.249~256 (2020)
- Park JH, "Automated identification of cephalometric landmarks : part 1-comparisons between the latest deep-learning methods YOLOV3 and SSD", Angle Orthod, 89, pp.903~909 (2019)
- Hwang HW, "Automated identification of cephalometric landmarks : part 2-might it be better than human?", Angle Orthod, 90, pp.69~76 (2020)
- Kunz F, "Artificial intelligence in orthodontics : evaluation of a fully automated cephalometric analysis using a customized convolutional neural network", J Orofac Orthop, 81, pp.52~68 (2020)
- Cephalometric analysis of normal occlusion in Korean adults, "Korean Association of Orthodontics Malocclusion White Paper Publication Committee", Korean Association of Orthodontists, ~ (1997)
- Bujang MA, "Guidelines of the minimum sample size requirements for Cohen’s Kappa", Epidemiol Biostat Public Health, 14, pp.e12267~ (2017)
- McHugh ML, "Interrater reliability: the Kappa statistic", Biochem Med(Zagreb), 22, pp.276~282 (2012)
- Russakovsky O, "ImageNet large scale visual recognition challenge", Int J Comput Vis, 115, pp.211~252 (2015)
- Huang G, "Densely connected convolutional networks", Paper presented at: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.2261~2269 (2017), IEEE
- "Accurate, large minibatch SGD: training ImageNet in 1 hour" (2018)[URL]
- "Highly scalable deep learning training system with mixed-precision: training ImageNet in four minutes" (2018)[URL]
- "Batch normalization: accelerating deep network training by reducing internal covariate shift" (2015)[URL]
- Wu Y, "ECCV 2018:Computer vision – ECCV 2018", pp.3~19 (2018), Springer
- Deng J, "ArcFace: additive angular margin loss for deep face recognition", Paper presented at: 2019 IEEE, 2019, 4690~4699[DOI]
- Dreiseitl S, "Comparing three-class diagnostic tests by three-way ROC analysis", Med Decis Making, 20, pp.323~331 (2000)
- Li J, "ROC analysis with multiple classes and multiple tests : methodology and its application in microarray studies", Biostatistics, 9, pp.566~576 (2008)
- van der Maaten L, "Visualizing data using t-SNE", J Mach Learn Res, 9, pp.2579~2605 (2008)
- Selvaraju RR, "Grad-CAM : visual explanations from deep networks via gradient-based localization", Paper presented at: 2017 IEEE International Conference on Computer Vision (ICCV), pp.618~626 (2017), IEEE