Imaging Science in Dentistry

Korean Academy of Oral and Maxillofacial Radiology (대한영상치의학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison of Multi-Label U-Net and Mask R-CNN for panoramic radiograph segmentation to detect periodontitis

  • Rini, Widyaningrum (Department of Dentomaxillofacial Radiology, Faculty of Dentistry, Universitas Gadjah Mada) ;
  • Ika, Candradewi (Department of Computer Science and Electronics, Faculty of Mathematics and Natural Sciences, Universitas Gadjah Mada) ;
  • Nur Rahman Ahmad Seno, Aji (Department of Periodontics, Faculty of Dentistry, Universitas Gadjah Mada) ;
  • Rona, Aulianisa (Faculty of Dentistry, Universitas Gadjah Mada)
  • Received : 2022.06.12
  • Accepted : 2022.09.09
  • Published : 2022.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

Purpose: Periodontitis, the most prevalent chronic inflammatory condition affecting teeth-supporting tissues, is diagnosed and classified through clinical and radiographic examinations. The staging of periodontitis using panoramic radiographs provides information for designing computer-assisted diagnostic systems. Performing image segmentation in periodontitis is required for image processing in diagnostic applications. This study evaluated image segmentation for periodontitis staging based on deep learning approaches. Materials and Methods: Multi-Label U-Net and Mask R-CNN models were compared for image segmentation to detect periodontitis using 100 digital panoramic radiographs. Normal conditions and 4 stages of periodontitis were annotated on these panoramic radiographs. A total of 1100 original and augmented images were then randomly divided into a training (75%) dataset to produce segmentation models and a testing (25%) dataset to determine the evaluation metrics of the segmentation models. Results: The performance of the segmentation models against the radiographic diagnosis of periodontitis conducted by a dentist was described by evaluation metrics(i.e., dice coefficient and intersection-over-union [IoU] score). MultiLabel U-Net achieved a dice coefficient of 0.96 and an IoU score of 0.97. Meanwhile, Mask R-CNN attained a dice coefficient of 0.87 and an IoU score of 0.74. U-Net showed the characteristic of semantic segmentation, and Mask R-CNN performed instance segmentation with accuracy, precision, recall, and F1-score values of 95%, 85.6%, 88.2%, and 86.6%, respectively. Conclusion: Multi-Label U-Net produced superior image segmentation to that of Mask R-CNN. The authors recommend integrating it with other techniques to develop hybrid models for automatic periodontitis detection.

Keywords

Acknowledgement

This work was funded by the Faculty of Dentistry Universitas Gadjah Mada under the grant "Penelitian Dana Masyarakat FKG UGM 2020" (Ref No. 3651/UN1/FKG1/Set.KG1/LT/2020).

References

  1. Papapanou PN, Sanz M, Buduneli N, Dietrich T, Feres M, Fine DH, et al. Periodontitis: consensus report of workgroup 2 of the 2017 World Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions. J Periodontol 2018; 89 Suppl 1: S173-82. https://doi.org/10.1002/JPER.17-0721
  2. Teeuw WJ, Coelho L, Silva A, van der Palen CJ, Lessmann FG, van der Velden U, et al. Validation of a dental image analyzer tool to measure alveolar bone loss in periodontitis patients. J Periodontal Res 2009; 44: 94-102. https://doi.org/10.1111/j.1600-0765.2008.01111.x
  3. Preshaw PM. Detection and diagnosis of periodontal conditions amenable to prevention. BMC Oral Health 2015; 15 Suppl 1: S5.
  4. Hanindriyo L, Widita E, Widyaningrum R, Priyono B, Agustina D. Influence of residential characteristics on the association between the oral health status and BMI of older adults in Indonesia. Gerodontology 2018; 35: 268-75. https://doi.org/10.1111/ger.12352
  5. Zia A, Hakim S, Khan AU, Bey A, Ateeq H, Parveen S, et al. Bone markers and bone mineral density associates with periodontitis in females with poly-cystic ovarian syndrome. J Bone Miner Metab 2022; 40: 487-97.
  6. Widita E, Hanindriyo L, Widyaningrum R, Priyono B, Agustina D. The association between periodontal conditions and serum lipids among elderly participants in Gadjah Mada medical centre, Yogyakarta. J Dent Indones 2017; 24: 63-9.
  7. Hosny A, Parmar C, Quackenbush J, Schwartz LH, Aerts HJ. Artificial intelligence in radiology. Nat Rev Cancer 2018; 18: 500-10. https://doi.org/10.1038/s41568-018-0016-5
  8. Gadosey PK, Li Y, Adjei Agyekum E, Zhang T, Liu Z, Yamak PT, et al. SD-UNet: stripping down U-Net for segmentation of biomedical images on platforms with low computational budgets. Diagnostics(Basel) 2020; 10: 110.
  9. Zhou Z, Siddiquee MM, Tajbakhsh N, Liang J. UNet++: redesigning skip connections to exploit multiscale features in image segmentation. IEEE Trans Med Imaging 2020; 39: 1856-67. https://doi.org/10.1109/tmi.2019.2959609
  10. Thanathornwong B, Suebnukarn S. Automatic detection of periodontal compromised teeth in digital panoramic radiographs using faster regional convolutional neural networks. Imaging Sci Dent 2020; 50: 169-74. https://doi.org/10.5624/isd.2020.50.2.169
  11. Alotaibi G, Awawdeh M, Farook FF, Aljohani M, Aldhafiri RM, Aldhoayan M. Artificial intelligence (AI) diagnostic tools: utilizing a convolutional neural network (CNN) to assess periodontal bone level radiographically-a retrospective study. BMC Oral Health 2022; 22: 399.
  12. Chang HJ, Lee SJ, Yong TH, Shin NY, Jang BG, Kim JE, et al. Deep learning hybrid method to automatically diagnose periodontal bone loss and stage periodontitis. Sci Rep 2020; 10: 7531.
  13. Jiang L, Chen D, Cao Z, Wu F, Zhu H, Zhu F. A two-stage deep learning architecture for radiographic staging of periodontal bone loss. BMC Oral Health 2022; 22: 106.
  14. Dev S, Manandhar S, Lee YH, Winkler S. Multi-label cloud segmentation using a deep network. In: 2019 USNC-URSI Radio Science Meeting (Joint with AP-S Symposium); 2019 Jul 7-12; Atlanta. IEEE; 2019: p. 113-4.
  15. He K, Gkioxari G, Dollar P, Girshick R. Mask R-CNN. IEEE Trans Pattern Anal Mach Intell 2020; 42: 386-97. https://doi.org/10.1109/TPAMI.2018.2844175
  16. Falk T, Mai D, Bensch R, Cicek O, Abdulkadir A, Marrakchi Y, et al. U-Net: deep learning for cell counting, detection, and morphometry. Nat Methods 2019; 16: 67-70. https://doi.org/10.1038/s41592-018-0261-2
  17. Loh R, Yong WX, Yapeter J, Subburaj K, Chandramohanadas R. A deep learning approach to the screening of malaria infection: automated and rapid cell counting, object detection and instance segmentation using Mask R-CNN. Comput Med Imaging Graph 2021; 88: 101845.
  18. Zhou LQ, Wang JY, Yu SY, Wu GG, Wei Q, Deng YB, et al. Artificial intelligence in medical imaging of the liver. World J Gastroenterol 2019; 25: 672-82. https://doi.org/10.3748/wjg.v25.i6.672
  19. Schwendicke F, Golla T, Dreher M, Krois J. Convolutional neural networks for dental image diagnostics: a scoping review. J Dent 2019; 91: 103226.
  20. Abdi AH, Kasaei S, Mehdizadeh M. Automatic segmentation of mandible in panoramic X-ray. J Med Imaging (Bellingham) 2015; 2: 044003.
  21. Shorten C, Khoshgoftaar TM. A survey on image data augmentation for deep learning. J Big Data 2019; 6: 0.
  22. Kim J, Lee HS, Song IS, Jung KH. DeNTNet: Deep Neural Transfer Network for the detection of periodontal bone loss using panoramic dental radiographs. Sci Rep 2019; 9: 17615.
  23. da Silva Rocha E, Endo PT. A comparative study of deep learning models for dental segmentation in panoramic radiograph. Appl Sci(Basel) 2022; 12: 3103.
  24. Kanuri N, Abdelkarim AZ, Rathore SA. Trainable WEKA (Waikato Environment for Knowledge Analysis) segmentation tool: machine-learning-enabled segmentation on features of panoramic radiographs. Cureus 2022; 14: e21777.
  25. Cantu AG, Gehrung S, Krois J, Chaurasia A, Rossi JG, Gaudin R, et al. Detecting caries lesions of different radiographic extension on bitewings using deep learning. J Dent 2020; 100: 103425.
  26. Lian L, Zhu T, Zhu F, Zhu H. Deep learning for caries detection and classification. Diagnostics(Basel) 2021; 11: 1672.
  27. Lee JH, Han SS, Kim YH, Lee C, Kim I. Application of a fully deep convolutional neural network to the automation of tooth segmentation on panoramic radiographs. Oral Surg Oral Med Oral Pathol Oral Radiol 2019; 129: 635-42.