Imaging Science in Dentistry

  • 제54권3호
  • /
  • Pages.257-263
  • /
  • 2024
  • /
  • 2233-7822(pISSN)
  • /
  • 2233-7830(eISSN)
대한영상치의학회 (Korean Academy of Oral and Maxillofacial Radiology)
권호 표지
DOI QR코드

DOI QR Code

Classification of mandibular molar furcation involvement in periapical radiographs by deep learning

  • Katerina Vilkomir (Department of Computer Science, East Carolina University) ;
  • Cody Phen (School of Dental Medicine, East Carolina University) ;
  • Fiondra Baldwin (School of Dental Medicine, East Carolina University) ;
  • Jared Cole (School of Dental Medicine, East Carolina University) ;
  • Nic Herndon (Department of Computer Science, East Carolina University) ;
  • Wenjian Zhang (School of Dental Medicine, East Carolina University)
  • 투고 : 2024.02.13
  • 심사 : 2024.06.26
  • 발행 : 2024.09.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

Purpose: The purpose of this study was to classify mandibular molar furcation involvement (FI) in periapical radiographs using a deep learning algorithm. Materials and Methods: Full mouth series taken at East Carolina University School of Dental Medicine from 2011-2023 were screened. Diagnostic-quality mandibular premolar and molar periapical radiographs with healthy or FI mandibular molars were included. The radiographs were cropped into individual molar images, annotated as "healthy" or "FI," and divided into training, validation, and testing datasets. The images were preprocessed by PyTorch transformations. ResNet-18, a convolutional neural network model, was refined using the PyTorch deep learning framework for the specific imaging classification task. CrossEntropyLoss and the AdamW optimizer were employed for loss function training and optimizing the learning rate, respectively. The images were loaded by PyTorch DataLoader for efficiency. The performance of ResNet-18 algorithm was evaluated with multiple metrics, including training and validation losses, confusion matrix, accuracy, sensitivity, specificity, the receiver operating characteristic (ROC) curve, and the area under the ROC curve. Results: After adequate training, ResNet-18 classified healthy vs. FI molars in the testing set with an accuracy of 96.47%, indicating its suitability for image classification. Conclusion: The deep learning algorithm developed in this study was shown to be promising for classifying mandibular molar FI. It could serve as a valuable supplemental tool for detecting and managing periodontal diseases.

키워드

참고문헌

  1. Hashimoto DA, Rosman G, Rus D, Meireles OR. Artificial intelligence in surgery: promises and perils. Ann Surg 2018; 268: 70-6. https://doi.org/10.1097/SLA.0000000000002693
  2. Wong SH, Al-Hasani H, Alam Z, Alam A. Artificial intelligence in radiology: how will we be affected? Eur Radiol 2019; 29: 141-3. https://doi.org/10.1007/s00330-018-5644-3
  3. Sklan JE, Plassard AJ, Fabbri D, Landman BA. Toward content based image retrieval with deep convolutional neural networks. Proc SPIE Int Soc Opt Eng 2015; 9417: 94172C.
  4. Thrall JH, Li X, Li Q, Cruz C, Do S, Dreyer K, et al. Artificial intelligence and machine learning in radiology: opportunities, challenges, pitfalls, and criteria for success. J Am Coll Radiol 2018; 15: 504-8. https://doi.org/10.1016/j.jacr.2017.12.026
  5. Liu Y, Balagurunathan Y, Atwater T, Antic S, Li Q, Walker RC, et al. Radiological image traits predictive of cancer status in pulmonary nodules. Clin Cancer Res 2017; 23: 1442-9. https://doi.org/10.1158/1078-0432.CCR-15-3102
  6. Schuhbaeck A, Otaki Y, Achenbach S, Schneider C, Slomka P, Berman DS, et al. Coronary calcium scoring from contrast coronary CT angiography using a semiautomated standardized method. J Cardiovasc Comput Tomogr 2015; 9: 446-53. https://doi.org/10.1016/j.jcct.2015.06.001
  7. Arimura H, Li Q, Korogi Y, Hirai T, Katsuragawa S, Yamashita Y, et al. Computerized detection of intracranial aneurysms for three-dimensional MR angiography: feature extraction of small protrusions based on a shape-based difference image technique. Med Phys 2006; 33: 394-401. https://doi.org/10.1118/1.2163389
  8. Wang S, Yao J, Summers RM. Improved classifier for computer-aided polyp detection in CT colonography by nonlinear dimensionality reduction. Med Phys 2008; 35: 1377-86. https://doi.org/10.1118/1.2870218
  9. Lee JH, Kim DH, Jeong SN, Choi SH. Detection and diagnosis of dental caries using a deep learning-based convolutional neural network algorithm. J Dent 2018; 77: 106-11. https://doi.org/10.1016/j.jdent.2018.07.015
  10. Lin PL, Huang PW, Huang PY, Hsu HC. Alveolar bone-loss area localization in periodontitis radiographs based on threshold segmentation with a hybrid feature fused of intensity and the H-value of fractional Brownian motion model. Comput Methods Programs Biomed 2015; 121: 117-26. https://doi.org/10.1016/j.cmpb.2015.05.004
  11. Lin PL, Huang PY, Huang PW. Automatic methods for alveolar bone loss degree measurement in periodontitis periapical radiographs. Comput Methods Programs Biomed 2017; 148: 1-11. https://doi.org/10.1016/j.cmpb.2017.06.012
  12. Okada K, Rysavy S, Flores A, Linguraru MG. Noninvasive differential diagnosis of dental periapical lesions in cone-beam CT scans. Med Phys 2015; 42: 1653-65. https://doi.org/10.1118/1.4914418
  13. Abdolali F, Zoroofi RA, Otake Y, Sato Y. Automated classification of maxillofacial cysts in cone beam CT images using contourlet transformation and spherical harmonics. Comput Methods Programs Biomed 2017; 139: 197-207. https://doi.org/10.1016/j.cmpb.2016.10.024
  14. Yilmaz E, Kayikcioglu T, Kayipmaz S. Computer-aided diagnosis of periapical cyst and keratocystic odontogenic tumor on cone beam computed tomography. Comput Methods Programs Biomed 2017; 146: 91-100. https://doi.org/10.1016/j.cmpb.2017.05.012
  15. Nibali L, Zavattini A, Nagata K, Di Iorio A, Lin GH, Needleman I, et al. Tooth loss in molars with and without furcation involvement - a systematic review and meta-analysis. J Clin Periodontol 2016, 43: 156-66.
  16. Goodson JM. Antimicrobial strategies for treatment of periodontal diseases. Periodontol 2000 1994; 5: 142-68. https://doi.org/10.1111/j.1600-0757.1994.tb00022.x
  17. McFall WT Jr. Tooth loss in 100 treated patients with periodontal disease. A long-term study. J Periodontol 1982; 53: 539-49. https://doi.org/10.1902/jop.1982.53.9.539
  18. Svardstrom G, Wennstrom JL. Prevalence of furcation involvements in patients referred for periodontal treatment. J Clin Periodontol 1996; 23: 1093-9. https://doi.org/10.1111/j.1600-051X.1996.tb01809.x
  19. Graetz C, Plaumann A, Wiebe JF, Springer C, Salzer S, Dorfer CE. Periodontal probing versus radiographs for the diagnosis of furcation involvement. J Periodontol 2014; 85: 1371-9. https://doi.org/10.1902/jop.2014.130612
  20. Shaker ZMH, Parsa A, Moharamzadeh K. Development of a radiographic index for periodontitis. Dent J(Basel) 2021; 9: 19.
  21. Abbas F, Hart AA, Oosting J, van der Velden U. Effect of training and probing force on the reproducibility of pocket depth measurements. J Periodontal Res 1982; 17: 226-34. https://doi.org/10.1111/j.1600-0765.1982.tb01149.x
  22. Bragger U. Radiographic parameters: biological significance and clinical use. Periodontol 2000 2005; 39: 73-90. https://doi.org/10.1111/j.1600-0757.2005.00128.x
  23. Eickholz P. Reproducibility and validity of furcation measurements as related to class of furcation invasion. J Periodontol 1995; 66: 984-9.
  24. Eickholz P, Hausmann E. Accuracy of radiographic assessment of interproximal bone loss in intrabony defects using linear measurements. Eur J Oral Sci 2000; 108: 70-3. https://doi.org/10.1034/j.1600-0722.2000.00729.x
  25. Muller HP, Eger T. Furcation diagnosis. J Clin Periodontol 1999; 26: 485-98.
  26. Mao YC, Huang YC, Chen TY, Li KC, Lin YJ, Liu YL, et al. Deep learning for dental diagnosis: a novel approach to furcation involvement detection on periapical radiographs. Bioengineering (Basel) 2023; 10: 802.
  27. Glick A, Clayton M, Angelov N, Chang J. Impact of explainable artificial intelligence assistance on clinical decision-making of novice dental clinicians. JAMIA Open 2022; 5: ooac031.
  28. Chlap P, Min H, Vandenberg N, Dowling J, Holloway L, Haworth A. A review of medical image data augmentation techniques for deep learning applications. J Med Imaging Radiat Oncol 2021; 65: 545-63. https://doi.org/10.1111/1754-9485.13261
  29. Graetz C, Schutzhold S, Plaumann A, Kahl M, Springer C, Salzer S, et al. Prognostic factors for the loss of molars - an 18-years retrospective cohort study. J Clin Periodontol 2015; 42: 943-50. https://doi.org/10.1111/jcpe.12460
  30. Reddy MS, Aichelmann-Reidy ME, Avila-Ortiz G, Klokkevold PR, Murphy KG, Rosen PS, et al. Periodontal regeneration - furcation defects: a consensus report from the AAP Regeneration Workshop. J Periodontol 2015; 86(2 Suppl): S131-3.
  31. Jolivet G, Huck O, Petit C. Evaluation of furcation involvement with diagnostic imaging methods: a systematic review. Dentomaxillofac Radiol 2022; 51: 20210529.
  32. Alasqah M, Alotaibi FD, Gufran K. The radiographic assessment of furcation area in maxillary and mandibular first molars while considering the new classification of periodontal disease. Healthcare (Basel) 2022; 10: 1464.