J Periodontal Implant Sci.  2018 Apr;48(2):114-123. 10.5051/jpis.2018.48.2.114.

Diagnosis and prediction of periodontally compromised teeth using a deep learning-based convolutional neural network algorithm

• Lee JH
• Kim DH
• Jeong SN
• Choi SH

• KMID: 2410690
• DOI: http://doi.org/10.5051/jpis.2018.48.2.114

Abstract

PURPOSE
The aim of the current study was to develop a computer-assisted detection system based on a deep convolutional neural network (CNN) algorithm and to evaluate the potential usefulness and accuracy of this system for the diagnosis and prediction of periodontally compromised teeth (PCT).
METHODS
Combining pretrained deep CNN architecture and a self-trained network, periapical radiographic images were used to determine the optimal CNN algorithm and weights. The diagnostic and predictive accuracy, sensitivity, specificity, positive predictive value, negative predictive value, receiver operating characteristic (ROC) curve, area under the ROC curve, confusion matrix, and 95% confidence intervals (CIs) were calculated using our deep CNN algorithm, based on a Keras framework in Python.
RESULTS
The periapical radiographic dataset was split into training (n=1,044), validation (n=348), and test (n=348) datasets. With the deep learning algorithm, the diagnostic accuracy for PCT was 81.0% for premolars and 76.7% for molars. Using 64 premolars and 64 molars that were clinically diagnosed as severe PCT, the accuracy of predicting extraction was 82.8% (95% CI, 70.1%-91.2%) for premolars and 73.4% (95% CI, 59.9%-84.0%) for molars.
CONCLUSIONS
We demonstrated that the deep CNN algorithm was useful for assessing the diagnosis and predictability of PCT. Therefore, with further optimization of the PCT dataset and improvements in the algorithm, a computer-aided detection system can be expected to become an effective and efficient method of diagnosing and predicting PCT.

Keyword

Artificial intelligence; Machine learning; Periodontal diseases; Supervised machine learning

MeSH Terms

Area Under Curve
Artificial Intelligence
Bicuspid
Boidae
Dataset
Diagnosis*
Learning
Machine Learning
Methods
Molar
Periodontal Diseases
ROC Curve
Sensitivity and Specificity
Supervised Machine Learning
Tooth*
Weights and Measures

Figure

• 

  Figure 1 Overall architecture of the deep CNN model. The dataset for the PCT images (224×224 pixels) is labeled as the input. Each of the convolutional layers is followed by a ReLU activation function, dropout, maximum pooling layers, and 3 fully connected layers with 1,024, 1,024, and 512 nodes, respectively. The final output layer performs 3 classifications using the Softmax function. CNN: convolutional neural network, PCT: periodontally compromised tooth, ReLU: rectified linear unit.

• 

  Figure 2 Multiclass classification confusion matrix with and without normalization using a deep CNN classifier. The diagonal elements are the number of points where the predicted label was the same as the actual label, while the non-diagonal elements were misinterpreted by the classifier. The higher the diagonal value and the darker the shade of blue, the more accurate the diagnosis of health and periodontally compromised teeth (A, B) Premolars without/with normalization. (C, D) Molars without/with normalization. CNN: convolutional neural network.

Reference

1. Tonetti MS, Jepsen S, Jin L, Otomo-Corgel J. Impact of the global burden of periodontal diseases on health, nutrition and wellbeing of mankind: a call for global action. J Clin Periodontol. 2017; 44:456–462.

2. Lee JH, Lee JS, Choi JK, Kweon HI, Kim YT, Choi SH. National dental policies and socio-demographic factors affecting changes in the incidence of periodontal treatments in Korean: A nationwide population-based retrospective cohort study from 2002-2013. BMC Oral Health. 2016; 16:118.

3. Lee JH, Oh JY, Choi JK, Kim YT, Park YS, Jeong SN, et al. Trends in the incidence of tooth extraction due to periodontal disease: results of a 12-year longitudinal cohort study in South Korea. J Periodontal Implant Sci. 2017; 47:264–272.

4. Lee JH, Lee JS, Park JY, Choi JK, Kim DW, Kim YT, et al. Association of lifestyle-related comorbidities with periodontitis: a nationwide cohort study in Korea. Medicine (Baltimore). 2015; 94:e1567.

5. Lee JH, Choi JK, Kim SH, Cho KH, Kim YT, Choi SH, et al. Association between periodontal flap surgery for periodontitis and vasculogenic erectile dysfunction in Koreans. J Periodontal Implant Sci. 2017; 47:96–105.

6. Lee JH, Oh JY, Youk TM, Jeong SN, Kim YT, Choi SH. Association between periodontal disease and non-communicable diseases: a 12-year longitudinal health-examinee cohort study in South Korea. Medicine (Baltimore). 2017; 96:e7398.

7. Choi JK, Kim YT, Kweon HI, Park EC, Choi SH, Lee JH. Effect of periodontitis on the development of osteoporosis: results from a nationwide population-based cohort study (2003–2013). BMC Womens Health. 2017; 17:77.

8. Lee JH, Kweon HH, Choi JK, Kim YT, Choi SH. Association between periodontal disease and prostate cancer: results of a 12-year longitudinal cohort study in South Korea. J Cancer. 2017; 8:2959–2965.

9. Graziani F, Karapetsa D, Alonso B, Herrera D. Nonsurgical and surgical treatment of periodontitis: how many options for one disease? Periodontol 2000. 2017; 75:152–188.

10. Martins SH, Novaes AB Jr, Taba M Jr, Palioto DB, Messora MR, Reino DM, et al. Effect of surgical periodontal treatment associated to antimicrobial photodynamic therapy on chronic periodontitis: a randomized controlled clinical trial. J Clin Periodontol. 2017; 44:717–728.

11. Ainamo J, Barmes D, Beagrie G, Cutress T, Martin J, Sardo-Infirri J. Development of the World Health Organization (WHO) community periodontal index of treatment needs (CPITN). Int Dent J. 1982; 32:281–291.

12. Sklan JE, Plassard AJ, Fabbri D, Landman BA. Toward content based image retrieval with deep convolutional neural networks. In : Proc SPIE Int Soc Opt Eng; 2015. p. 9417.

13. Rajpurkar P, Irvin J, Zhu K, Yang B, Mehta H, Duan T, et al. CheXNet: radiologist-level pneumonia detection on chest X-rays with deep learning. arXiv e-print 2017;arXiv:1711.05225.

14. Rajpurkar P, Hannun AY, Haghpanahi M, Bourn C, Ng AY. Cardiologist-level arrhythmia detection with convolutional neural networks. arXiv e-print 2017;arXiv:1707.01836.

15. Garcia-Hernandez JJ, Gomez-Flores W, Rubio-Loyola J. Analysis of the impact of digital watermarking on computer-aided diagnosis in medical imaging. Comput Biol Med. 2016; 68:37–48.

16. Litjens G, Kooi T, Bejnordi BE, Setio AA, Ciompi F, Ghafoorian M, et al. A survey on deep learning in medical image analysis. Med Image Anal. 2017; 42:60–88.

17. Kim TS, Obst C, Zehaczek S, Geenen C. Detection of bone loss with different X-ray techniques in periodontal patients. J Periodontol. 2008; 79:1141–1149.

18. Armitage GC. Periodontal diagnoses and classification of periodontal diseases. Periodontol 2000. 2004; 34:9–21.

19. Page RC, Eke PI. Case definitions for use in population-based surveillance of periodontitis. J Periodontol. 2007; 78:1387–1399.

20. Shin HC, Roth HR, Gao M, Lu L, Xu Z, Nogues I, et al. Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning. IEEE Trans Med Imaging. 2016; 35:1285–1298.

21. Ohsugi H, Tabuchi H, Enno H, Ishitobi N. Accuracy of deep learning, a machine-learning technology, using ultra-wide-field fundus ophthalmoscopy for detecting rhegmatogenous retinal detachment. Sci Rep. 2017; 7:9425.

22. Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. arXiv e-print 2014;arXiv:1409.556.

23. Nair V, Hinton GE. Rectified linear units improve restricted boltzmann machines. In : Proceedings of the 27th International Conference on International Conference on Machine Learning; 2010 Jun 21–24; Haifa. Madison (WI): Omnipress;2010. p. 807–814.

24. Chollet F. Keras [Internet]. San Francisco (CA): GitHub, Inc.;2017. cited 2018 Mar 19. Available from: https://github.com/keras-team/keras.

25. Abadi M, Agarwal A, Barham P, Brevdo E, Chen Z, Citro C, et al. TensorFlow: large-scale machine learning on heterogeneous distributed systems. arXiv e-print 2016;arXiv:1603.04467.

26. Lehman CD, Wellman RD, Buist DS, Kerlikowske K, Tosteson AN, Miglioretti DL. Diagnostic accuracy of digital screening mammography with and without computer-aided detection. JAMA Intern Med. 2015; 175:1828–1837.

27. Gulshan V, Peng L, Coram M, Stumpe MC, Wu D, Narayanaswamy A, et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA. 2016; 316:2402–2410.

28. Lakhani P, Sundaram B. Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks. Radiology. 2017; 284:574–582.

29. Wang R. Edge detection using convolutional neural network. In : In : Cheng L, Liu Q, Ronzhin A, editors. Advances in neural networks, ISNN 2016. 13th International Symposium on Neural Networks, ISNN 2016; 2016 Jul 6–8; Saint Petersburg. Cham: Springer International Publishing;2016. p. 12–20.

30. Ouyang W, Wang X. Joint deep learning for pedestrian detection. In : 2013 IEEE International Conference on Computer Vision (ICCV); 2013 Dec 1–8; Sydney. Piscataway (NJ): IEEE;2013. p. 2056–2063.

31. Szegedy C, Vanhoucke V, Ioffe S, Shlens J, Wojna Z. Rethinking the inception architecture for computer vision. In : The IEEE Conference on Computer Vision and Pattern Recognition (CVPR); 2016 Jun 26–Jul 1; Las Vegas Valley (NV). Piscataway (NJ): IEEE;2016. p. 2818–2826. .

32. Keskar NS, Mudigere D, Nocedal J, Smelyanskiy M, Tang PT. On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima. arXiv e-print 2017;arXiv:1609.0483.

33. Esteva A, Kuprel B, Novoa RA, Ko J, Swetter SM, Blau HM, et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature. 2017; 542:115–118.

34. Peng X, Sun B, Ali K, Saenko K. Learning deep object detectors from 3D models. In : 2015 IEEE International Conference on Computer Vision (ICCV); 2015 Dec 7–13; Santiago. Piscataway (NJ): IEEE;2015. p. 1278–1286.