Impacted Tooth Detection and Segmentation Based on Deep Convolutional Neural Network in Panoramic Dental Images

Öz

Impacted tooth detection is an important step in dental practice and an accurate fixation process is of great importance in treatment planning and diagnosis. Considering the limitations and error probabilities of traditional methods, the utilization of artificial intelligence-based approaches like deep learning models is becoming more common. In this study, the performance of deep learning models was evaluated for the detection and segmentation of impacted teeth in panoramic dental images. The performance of seven different models (VGG16-Unet, VGG19-Unet, MobileNetV2, Unet-v1, Unet-v2, Unet-v3 and Unet-v4) was evaluated. The AUC (area under the curve) value of the VGG16-Unet model was found to be higher with 94.87% compared to the other models. This study contributes to the advancement of more accurate and sensitive segmentation methods in the field of dentistry and supports more reliable results in tooth detection and treatment planning processes.

Anahtar Kelimeler

• Unet
• Deep learning
• Impacted tooth
• Panoramic dental images

Panoramik Diş Görüntülerinde Derin Evrişimsel Sinir Ağına Dayalı Gömülü Diş Tespiti ve Segmentasyonu

Öz

Gömülü diş tespiti, diş hekimliği uygulamalarında önemli bir adımdır ve doğru bir tespit süreci, tedavi planlaması ve teşhislerde büyük önem taşır. Geleneksel yöntemlerin sınırlamaları ve hata olasılıkları göz önüne alındığında, derin öğrenme modelleri gibi yapay zekâ temelli yaklaşımların kullanılması giderek daha yaygın hale gelmektedir. Bu çalışmada panoramik gömülü diş görüntülerinde derin öğrenme modellerinin performansı incelenmiştir Yedi farklı modelin (VGG16-Unet, VGG19-Unet, MobileNetV2, Unet-v1, Unet-v2, Unet-v3 ve Unet-v4) performansı değerlendirilmiştir. VGG16-Unet modelinin AUC (eğri altındaki alan) değeri %94.87 ile diğer modellere kıyasla daha yüksek bulunmuştur. Bu çalışma, diş hekimliği alanında daha doğru ve hassas segmentasyon yöntemleri geliştirilmesine katkı sağlayarak, diş tespiti ve tedavi planlaması süreçlerinde daha güvenilir sonuçlar elde edilmesini desteklemektedir.

Anahtar Kelimeler

• Unet
• Derin öğrenme
• Gömülü diş
• Panoramik diş görüntüleri

Kaynakça

1. 1. Narayan, V., Mall, P.K., Alkhayyat, A., Abhishek, K., Kumar, S., Pandey, P., 2023. Enhance-Net: An Approach to Boost the Performance of Deep Learning Model Based on Real-Time Medical Images. Journal of Sensors,15.
2. 2. Kumar, A., Nag, A., Jain, N., Bandopadhyay, S., 2023. Surgical Management of Impacted Canine: A Case Series. Journal of Dental Health & Research (JDHR), 4(1), 11-15.
3. 3. Tetay-Salgado, S., Arriola-Guillén, L.E., Ruíz-Mora, G.A., Aliaga-Del Castillo, A., Rodríguez-Cárdenas, Y.A., 2021. Prevalence of Impacted Teeth and Supernumerary Teeth by Radiographic Evaluation in Three Latin American Countries: A Cross-Sectional Study. Journal of Clinical and Experimental Dentistry, 13(4), 363-368.
4. 4. Singh, N.K., Raza, K., 2022. Progress in Deep Learning-Based Dental and Maxillofacial İmage Analysis: A Systematic Review. Expert Systems with Applications, 199, 116968, 15.
5. 5. Lee, J.H., Kim, D.H., Jeong, S.N., Choi, S.H., 2018. Detection and Diagnosis of Dental Caries using A Deep Learning-Based Convolutional Neural Network Algorithm. Journal of Dentistry, 77, 106-111.
6. 6. Elborolosy, S.A., Salem, W.S., Hamed, M.O., Sayed, A.S., Helmy, B.E.D., Elngar, A.A., 2022. Predicting Difficulty Level of Surgical Removal of Impacted Mandibular Third Molar using Deep Learning Approaches. Research Square, 21.
7. 7. Krois, J., Schneider, L., Schwendicke, F., 2021. Impact of Image Context on Deep Learning for Classification of Teeth on Radiographs. Journal of Clinical Medicine, 10(8), 1635, 9.
8. 8. Tuzoff, D.V., Tuzova, L.N., Bornstein, M.M., Krasnov, A.S., Kharchenko, M.A., Nikolenko, S.I., Sveshnikov, M.M., Bednenko, G.B., 2019. Tooth Detection and Numbering in Panoramic Radiographs using Convolutional Neural Networks. Dentomaxillofacial Radiology (DMFR), 48(4), 20180051, 15.

9. 9. Hiraiwa, T., Ariji, Y., Fukuda, M., Kise, Y., Nakata, K., Katsumata, A., Fujita, H., Ariji, E., 2019. A Deep-Learning Artificial İntelligence System for Assessment of Root Morphology of the Mandibular First Molar on Panoramic Radiography. Dentomaxillofacial Radiology (DMFR), 48(3), 20180218, 7.
10. 10. Imak, A., Celebi, A., Polat, O., Turkoglu, M., Sengur, A., 2023. ResMIBCU-Net: An Encoder-Decoder Network with Residual Blocks, Modified İnverted Residual Block, and Bi-Directional ConvLSTM for Impacted Tooth Segmentation in Panoramic X-Ray İmages. Oral Radiology, 1, 1-15.
11. 11. Román, J.C.M., Fretes, V.R., Adorno, C.G., Silva, R.G., Noguera, J.L.V., Legal-Ayala, H., Mello-Román, J.D., Torres, R.D.E., Facon, J., 2021. Panoramic Dental Radiography Image Enhancement using Multiscale Mathematical Morphology. Sensors, 21(9), 3110, 19.
12. 12. Ronneberger, O., Fischer, P., Brox, T., 2015. U-net: Convolutional Networks for Biomedical Image Segmentation, Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, 9351, 234-241.
13. 13. Balakrishna, C., Dadashzadeh, S., Soltaninejad, S., 2018. Automatic Detection of Lumen and Media in the IVUS Images using U-Net with VGG16 Encoder. arXiv preprint arXiv:1806.07554, 10.
14. 14. Ali, R., Hardie, R.C., Narayanan, B.N., De Silva, S., 2019. Deep Learning Ensemble Methods for Skin Lesion Analysis Towards Melanoma Detection. 2019 IEEE National Aerospace and Electronics Conference (NAECON), Dayton, 311-316.
15. 15. Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., Chen, L.C., 2018. Mobilenetv2: Inverted Residuals and Linear Bottlenecks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, 4510-4520.
16. 16. Sokolova, M., Lapalme, G., 2009. A Systematic Analysis of Performance Measures For Classification Tasks. Information Processing & Management, 45(4), 427-437.
17. 17. Theckedath, D., Sedamkar, R.R., 2020. Detecting Affect States using VGG16, ResNet50 and SE-ResNet50 Networks, SN Computer Science, 1(2), 79, 7.
18. 18. Cheng, D., Lam, E.Y., 2021. Transfer learning U-Net Deep Learning for Lung Ultrasound Segmentation. arXiv preprint arXiv: 2110.02196, 14.
19. 19. Salama, W.M., Aly, M.H., 2021. Deep Learning in Mammography Images Segmentation and Classification: Automated CNN Approach. Alexandria Engineering Journal, 60(5), 4701-4709.
20. 20. Basaran, M., Celik, O., Bayrakdar, I.S., Bilgir, E., Orhan, K., Odabas, A., Arslan, A.F., Jagtap, R., 2022. Diagnostic Charting of Panoramic Radiography using Deep-Learning Artificial İntelligence System. Oral Radiology, 38, 363-369.
21. 21. Celik, O., 2021. Detection of Impacted Teeth using Deep Learning Technique. Comptes Rendus de L'Academie Bulgare des Sciences, 74(2), 269-277.
22. 22. Orhan, K., Bilgir, E., Bayrakdar, I.S., Ezhov, M., Gusarev, M., Shumilov, E., 2021. Evaluation of Artificial Intelligence for Detecting Impacted Third Molars on Cone-Beam Computed Tomography Scans. Journal of Stomatology, Oral and Maxillofacial Surgery, 122(4), 333-337.
23. 23. Celik, M.E., 2022. Deep Learning based Detection Tool for Impacted Mandibular Third Molar Teeth, Diagnostics, 12(4), 942, 31-43, 13.
24. 24. Kim, J.Y., Kahm, S.H., Yoo, S., Bae, S.M., Kang, J.E., Lee, S.H., 2023. The Efficacy of Supervised Learning and Semi-Supervised Learning in Diagnosis of Impacted Third Molar on Panoramic Radiographs Through Artificial Intelligence Model. Dentomaxillofacial Radiology (DMFR), 52(6), 12.