THE DEEP LEARNING MODEL FOR DECAYED-MISSING-FILLED TEETH DETECTION: A COMPARISON BETWEEN YOLOV5 AND YOLOV8

(Received: 19-Mar.-2024, Revised: 17-May-2024 , Accepted: 6-Jul.-2024)

Authors Maya Fitria, Yasmina Elma, Maulisa Oktiana *, Khairun Saddami, Rizki Novita, Rizkika Putri, Handika Rahayu, Hafidh Habibie, Subhan Janura,

Keywords #Caries detection #Detection model #Deep learning #DMF-T #Tooth decay

Abstract Tooth decay is a dental condition characterized by the deterioration of tooth tissue originating from the outer surface and progressing to the pulp. Severe tooth decay, evolving into cavities, necessitates timely intervention to avert more serious dental-health issues. Common treatment procedures include filling and extraction of affected teeth. Presently, dentists conduct examinations for tooth decay by manually tallying affected, missing and filled teeth using an odontogram—a human tooth code diagram. This data is then recorded in patients' dental medical records. Recognizing the need for automation in assessing patients' experiences of tooth decay, this research endeavors to develop a model capable of detecting decayed, missing and filled teeth using variations of the YOLOv5 and YOLOv8 model architectures. The results of the training evaluation demonstrate the efficacy of YOLOv5l with a learning rate of 10-2, exhibiting a high precision value of 0.97, a recall of 0.858 and a mean average precision (mAP) of 0.904 within 1 hour and 18 minutes. According to the curves obtained in the training process, YOLOv5l shows great performance on the dental caries dataset, but precautions like early stopping are needed for a reliable and generalizable model. In contrast, YOLOv8 offers better training stability and larger variants perform better on the dental caries dataset, improving detection capabilities with continued training epochs.

References

[1] S. Prabhu, A. B. Acharya and M. V. Muddapur, "Are Teeth Useful in Estimating Stature?," Journal ofForensic and Legal Medicine, vol. 20, no. 5, pp. 460–464, DOI: 10.1016/j.jflm.2013.02.004, 2013.

[2] G. T. Huang, "Pulp and Dentin Tissue Engineering and Regeneration: Current Progress," RegenerativeMedicine, vol. 4, no. 5, pp. 697–707, DOI: 10.2217/rme.09.45, 2009.

[3] X. Huang et al., "Microenvironment Influences Odontogenic Mesenchymal Stem Cells Mediated DentalPulp Regeneration," Frontiers in Physiology, vol. 12, p. 656588, 2021.

[4] L. Cheng et al., "Expert Consensus on Dental Caries Management," Int. Journal of Oral Science, vol.14, no. 1, p. 17, 2022.

[5] H. Nilsson, J. Sanmartin Berglund and S. Renvert, "Longitudinal Evaluation of Periodontitis and ToothLoss among Older Adults," Journal of Clinical Periodontology, vol. 46, no. 10, pp. 1041–1049, 2019.

[6] M. J. Y. Yon, S. S. Gao, K. J. Chen, D. Duangthip, E. C. M. Lo and C. H. Chu, "MedicalModel in CariesManagement," Dentistry Journal, vol. 7, no. 2, p. 37, 2019.

[7] T. Kikuiri, K. Saito, A. Iida, Y. Yoshimura, Y. Yawaka and T. Shirakawa, "Occurrence of SubcutaneousEmphysema during a Caries Filling Procedure: A Case Report," Pediatric Dental Journal, vol. 32, no. 3, pp. 211–215, 2022.

[8] M. Zhou, J. Dong, L. Zha and Y. Liao, "Causal Association between Periodontal Diseases andCardiovascular Diseases," Genes, vol. 13, no. 1, p. 13, 2021.

[9] J.-H. Lee, D.-H. Kim, S.-N. Jeong and S.-H. Choi, "Detection and Diagnosis of Dental Caries Using aDeep Learning-based Convolutional Neural Network Algorithm," Journal of Dentistry, vol. 77, pp. 106–111, 2018.

[10] M. Rathee and A. Sapra, "Dental Caries," [Online], Available: https://www.ncbi.nlm.nih.gov/books/NBK51699/, 2019.

[11] Balitbangkes, "National Report of Basic Health Research 2018 (in Indonesian)," Indonesian Ministryof Health, p. 206, 2019. [Online], Available: https://repository.badankebijakan.kemkes.go.id/id/eprint/3514/1/Laporan%20Riskesdas%202018%20Nasional.pdf, [Accessed: February 2024].

[12] N. D. Ardiyanti, R. Adhani and I. Hatta, "Correlation between DMF-T Caries Index, Consumption ofDrinking Water and Tooth Brushing Behavior in Indonesian Communities (in Indonesian)," Dentin, vol. 6, no. 1, 2022.

[13] E. S. Wardhana, S. Suryono, A. Hernawan and L. E., Nugroho, "Evaluation of Format and Security ofDental Electronic Medical Record Systems in General Hospital Based on Legislation," Odonto: Dental Journal, vol. 9, Special Issue 1, pp. 80-89, 2022.

[14] G. Moradi, A. M. Bolbanabad, A. Moinafshar, H. Adabi, M. Sharafi and B. Zareie, "Evaluation of OralHealth Status Based on the Decayed, Missing and Filled Teeth (DMFT) Index," Iranian Journal of Public Health, vol. 48, no. 11, p. 2050, 2019.

[15] A. Alami, S. Erfanpoor, E. Lael-Monfared, A. Ramezani and A. Jafari, "Investigation of Dental CariesPrevalence, Decayed, Missing and Filled Teeth (DMF-T and DMF-T Indices) and the Associated Factors among 9-11 Years Old Children," Research Square, pp. 1-18, DOI: 10.21203/rs.2.21545/v1, 2020.

[16] J. A. Daza-Cardona, J. Vargas-Ramírez and M. A. Guapacha-Sánchez, "Doing Odontograms andDentists in the Classroom. Materiality and Affect in Dental Education," Tapuya: LatinAmerican Science, Technology and Society, vol. 4, no. 1, p. 1968635, 2021.

[17] E. D. Fadhillah et al., "Smart Odontogram: Dental Diagnosis of Patients Using Deep Learning," Proc.of 2021 IEEE Int. Electronics Symposium (IES), pp. 532-537, DOI: 10.1109/IES53407.2021.9594027, Surabaya, Indonesia, 2021.

[18] I. S. Bayrakdar et al., "Deep-learning Approach for Caries Detection and Segmentation on DentalBitewing Radiographs," Oral Radiology, vol. 38, no. 4, pp. 468-479, pp. 1–12, 2021.

[19] Y. Al-Hadeethi, M. Sayyed, H. Mohammed and L. Rimondini, "X-ray Photons AttenuationCharacteristics for Two Tellurite Based Glass Systems at Dental Diagnostic Energies," Ceramics International, vol. 46, no. 1, pp. 251–257, 2020.

[20] M. Fitria et al., "Development of Intraoral Clinical Image Dataset for Deep Learning Caries Detection,"Proc. of the 2023 IEEE 2nd Int. Conf. on Computer System, Information Technology and Electrical Engineering (COSITE), pp. 194–198, DOI: 10.1109/COSITE60233.2023.10249428, Banda Aceh, Indonesia, 2023.

[21] L. Que et al., "Prevalence of Dental Caries in the First Permanent Molar and Associated Risk Factorsamong Sixth-grade Students in São Tomé Island," BMC Oral Health, vol. 21, no. 1, pp. 1–10, 2021.

[22] K. Kim, K. Kim and S. Jeong, "Application of YOLO v5 and v8 for Recognition of Safety Risk Factorsat Construction Sites," Sustainability, vol. 15, no. 20, p. 15179, 2023.

[23] T. Diwan, G. Anirudh and J. V. Tembhurne, "Object Detection Using YOLO: Challenges, ArchitecturalSuccessors, Datasets and Applications," Multimedia Tools and Applications, vol. 82, no. 6, pp. 9243–9275, 2023.

[24] V. D. Matta et al., "Single Use Plastic Bottle Recognition and Classification Using Yolo V5 and V8Architectures," Proc. of the Int. Conf. on Cognitive Computing and Cyber Physical Systems, Part of the Book Series: Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, vol. 537, pp. 99– 106, Springer, 2023.

[25] E. Casas, L. Ramos, E. Bendek and F. Rivas-Echeverría, "Assessing the Effectiveness of YOLOArchitectures for Smoke and Wildfire Detection," IEEE Access, vol. 11, pp. 96554 – 96583, DOI: 10.1109/ACCESS.2023.3312217, 2023.

[26] J. Terven, D.-M. Córdova-Esparza and J.-A. Romero-González, "A Comprehensive Review of YOLOArchitectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-nas," Machine Learning and Knowledge Extraction, vol. 5, no. 4, pp. 1680–1716, 2023.

[27] I. P. Sary, S. Andromeda and E. U. Armin, "Performance Comparison of YOLOv5 and YOLOv8Architectures in Human Detection using Aerial Images," Ultima Computing: Jurnal Sistem Komputer, vol. 15, no. 1, pp. 8–13, 2023.

[28] Q. Lin, G. Ye, J. Wang and H. Liu, "RoboFlow: A Data-centric Workflow Management System forDeveloping AI-enhanced Robots," Proc. of the 5th Conf. on Robot Learning, vol. 164, pp. 1789–1794, [Online], Available: https://proceedings.mlr.press/v164/lin22c.html, PMLR, 2022.

[29] Y. Lee, J. Choi and K. Jo, "VSNet: Vehicle State Classification for Drone Image with MosaicAugmentation and Soft-label Assignment," Proc. of the Asian Conference on Intelligent Information and Database Systems, Part of the Book Series: Lecture Notes in Computer Science, vol. 13995, pp. 109–120, 2023.

[30] F. Dadboud, V. Patel, V. Mehta, M. Bolic and I. Mantegh, "Single-stage UAV Detection andClassification with YOLOv5: Mosaic Data Augmentation and PANet," Proc. of the 2021 17th IEEE Int. Conf. on Advanced Video and Signal Based Surveillance (AVSS), pp. 1–8, DOI: 10.1109/AVSS52988.2021.9663841, Washington, USA, 2021.

[31] T. B. Pun, A. Neupane, R. Koech and K.Walsh, "Detection and Counting of Root-knot Nematodes UsingYOLO Models with Mosaic Augmentation," Biosensors and Bioelectronics: X, vol. 15, p.100407, DOI: 10.1016/j.biosx.2023.100407, 2023.

[32] Q. Song et al., "Object Detection Method for Grasping Robot Based on Improved YOLOv5,"Micro-machines, vol. 12, no. 11, p. 1273, 2021.

[33] L. Wang et al., "Investigation into Recognition Algorithm of Helmet Violation Based on YOLOv5-CBAM- DCN," IEEE Access, vol. 10, pp. 60622–60632, 2022.

[34] W. Sheng et al., "Symmetry-based Fusion Algorithm for Bone Age Detection with YOLOv5 andResNet34," Symmetry, vol. 15, no. 7, p. 1377, 2023.

[35] B. Selcuk and T. Serif, "A Comparison of YOLOv5 and YOLOv8 in the Context of Mobile UIDetection," Proc. of the Int. Conf. on Mobile Web and Intelligent Information Systems, Part of the Book Series: Lecture Notes in Computer Science, vol. 13977, pp. 161– 174, Springer, 2023.

[36] G. Wen, M. Li, Y. Luo, C. Shi and Y. Tan, "The Improved YOLOv8 Algorithm Based on EMSPConvand SPE-head Modules," Multimedia Tools and Applications, vol. 83, pp. 61007–61023, DOI: 10.1007/s11042-023-17957-4, 2024.

[37] H. Wang, C. Liu, Y. Cai, L. Chen and Y. Li, "YOLOv8-QSD: An Improved Small Object DetectionAlgorithm for Autonomous Vehicles Based on YOLOv8," IEEE Transactions on Instrumentation and Measurement, vol. 73, DOI: 10.1109/TIM.2024.3379090, 2024.