Journals / Düzce Üniversitesi Bilim ve Teknoloji Dergisi / 2022 / Cilt: 10 - Sayı: 1

Tooth Instance Segmentation on Panoramic Dental Radiographs Using U-Nets and Morphological Processing

U-Net ve Morfolojik İşlemler Kullanılarak Panoramik Radyografiler Üzerinde Diş Örneği Bölütleme

Pages
39–50
Publication Date
2022-10-12
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

A dice overlap score of 95.4±0.3% is obtained in overall teeth segmentation.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Automatic teeth segmentation in panoramic x-ray images is an important research subject of the image analysis in dentistry.
📊
Key Findings & Data:In this study, we propose a post-processing stage to obtain a segmentation map in which the objects in the image are separated, and apply this technique to tooth instance segmentation with U-Net network.
💡
Conclusion & Impact:A dice overlap score of 95.4±0.3% is obtained in overall teeth segmentation.
🔬Design:Research Article
📍Scope:Comprehensive Sample
⚙️Variables:Diş · Panoramik · Bölütleme

Abstract

Automatic teeth segmentation in panoramic x-ray images is an important research subject of the image analysis in dentistry. In this study, we propose a post-processing stage to obtain a segmentation map in which the objects in the image are separated, and apply this technique to tooth instance segmentation with U-Net network. The post-processing consists of grayscale morphological and filtering operations, which are applied to the sigmoid output of the network before binarization. A dice overlap score of 95.4±0.3% is obtained in overall teeth segmentation. The proposed post-processing stages reduce the mean error of tooth count to 6.15%, whereas the error without post-processing is 26.81%. The performances of both segmentation and tooth counting are the highest in the literature, to our knowledge. Moreover, this is achieved by using a relatively small training dataset, which consists of 105 images. Although the aim in this study is to segment tooth instances, the presented method is applicable to similar problems in other domains, such as separating the cell instances.

Özet

Panoramik röntgen görüntülerinde otomatik diş bölütleme, diş hekimliği görüntü analizinin önemli bir araştırma konusudur. Bu çalışmada, görüntüdeki nesnelerin ayrıldığı bir bölütleme haritası elde etmek için bir son işleme aşaması öneriyoruz ve bu tekniği U-Net ağı ile diş örneği bölütlemeye uyguluyoruz. Son işleme, ikilileştirmeden önce ağın sigmoid çıkışına uygulanan gri tonlamalı morfolojik ve filtreleme işlemlerinden oluşmaktadır. Tüm diş bölütlemede %95,4±0,3'lük bir Dice örtüşme puanı elde edilmiştir. Önerilen son işleme aşamaları, diş sayısının tespitinde ortalama hatayı %26,81'den %6,15'e düşürmüştür. Bildiğimiz kadarıyla hem bölütleme, hem de diş sayma performansları literatürdeki en yüksek performanslardır. Ayrıca bu sonuç, 105 görüntüden oluşan nispeten küçük bir eğitim veri seti kullanılarak elde edilmiştir. Bu çalışmadaki amaç diş örneklerini bölütlemek olsa da, sunulan yöntem hücre örneklerini ayırmak gibi diğer alanlardaki benzer problemlere uygulanabilir.

🔖

Cite This Article

HELLİ, S. S., HAMAMCI, A. (2022). Tooth Instance Segmentation on Panoramic Dental Radiographs Using U-Nets and Morphological Processing. Düzce Üniversitesi Bilim ve Teknoloji Dergisi, Cilt: 10 - Sayı: 1, 39–50. https://ajindex.com/articles/493838

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (1)

  1. M. Glick, D. M. Williams, D. V. Kleinman, M. Vujicic, R. G. Watt, and R. J. Weyant, “A new definition for oral health developed by the fdi world dental federation opens the door to a universal definition of oral health,” British dental journal, vol. 221, no. 12, pp. 792–793, 2016. [2]M. A. Peres, L. M. Macpherson, R. J. Weyant, B. Daly, R. Venturelli, M. R. Mathur, S. Listl, R. K. Celeste, C. C. Guarnizo-Herreño, C. Kearns et al., “Oral diseases: a global public health challenge,” The Lancet, vol. 394, no. 10194, pp. 249–260, 2019. [3]N. Shah, N. Bansal, and A. Logani, “Recent advances in imaging technologies in dentistry,” World journal of radiology, vol. 6, no. 10, p. 794, 2014. [4]B. Vandenberghe, R. Jacobs, and H. Bosmans, “Modern dental imaging: a review of the current technology and clinical applications in dental practice,” European radiology, vol. 20, no. 11, pp. 2637–2655, 2010. [5]R. G. Birdal, E. Gumus, A. Sertbas, and I. S. Birdal, “Automated lesion detection in panoramic dental radiographs,” Oral Radiology, vol. 32, no. 2, pp. 111–118, 2016. [6]E. Avuclu and F. Basciftci, “Novel approaches to determine age and gender from dental x-ray images by using multiplayer perceptron neural networks and image processing techniques,” Chaos, Solitons & Fractals, vol. 120, pp. 127 – 138, 2019. [7]M. H. Bozkurt and S. Karagol, “Jaw and teeth segmentation on the panoramic x-ray images for dental human identification,” Journal of Digital Imaging, pp. 1–18, 2020. [8]G. Jader, J. Fontineli, M. Ruiz, K. Abdalla, M. Pithon, and L. Oliveira, “Deep instance segmentation of teeth in panoramic x-ray images,” in 2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI). IEEE, 2018, pp. 400–407. [9]G. Silva, L. Oliveira, and M. Pithon, “Automatic segmenting teeth in x-ray images: Trends, a novel data set, benchmarking and future perspectives,” Expert Systems with Applications, vol. 107, pp. 15–31, 2018. [10]G. Litjens, T. Kooi, B. E. Bejnordi, A. A. A. Setio, F. Ciompi, M. Ghafoorian, J. A. Van Der Laak, B. Van Ginneken, and C. I. Sánchez, “A survey on deep learning in medical image analysis,” Medical image analysis, vol. 42, pp. 60–88, 2017. [11]Y. Zhao, P. Li, C. Gao, Y. Liu, Q. Chen, F. Yang, and D. Meng, “Tsasnet: Tooth segmentation on dental panoramic x-ray images by two-stage attention segmentation network,” Knowledge-Based Systems, vol. 206, p. 106338, 2020. [12]S. Ren, K. He, R. Girshick, and J. Sun, “Faster r-cnn: Towards real-time object d