Abstract:Dental plaque is one of the main etiologic factors that lead to dental caries and periodontal disease, and the amount of plaque on teeth could be served as reference for the teeth health condition to some extent. Therefore, the detection of dental plaque plays a very important role for maintaining the oral health. However, identification of dental plaque is difficult for both patient and dentist because the tooth and dental plaque often look similar, especially when plaque is present in scanty amounts. Excited by the light source with the wavelength of 405 nm, the auto-fluorescence effect will appear in both dental plaque and dental tissue, but the auto fluorescence spectrum of dental tissue mainly locates at the wavelength range of blue and green light, while the auto-fluorescence spectrum of dental plaque mainly locates at the wavelength range of red light, and the spectrum intensity caused by the different leveled dental plaque are also diverse. Based on the differences between auto-fluorescence spectra of dental plaque and dental tissues, a potable auto-fluorescence color imaging dental plaque detection system was developed. In the detection system, five surface-mounted LEDs whose central wavelength are all 405 nm are assembled as the excitation source, besides, a long pass optical filter with central wavelength of 520 nm is configured to improve the signal to noise ratio (SNR), and the excited auto-fluorescence was collected and imaged through the imaging lens to the array sensor of a color CCD with a resolution of 640 X 480. Finally, the amount of dental plaque is analyzed by processing the captured fluorescence images. An experiment was designed to confirm the reliability of the detection system. The anterior teeth auto-fluore scence images of seven recruited subjects with various amounts of dental plaque were captured by the system, after the subjects chewed the disclosing agent for 1 minute and rinsed their mouth with 100 mL distilled water, the Quigley Hein plaque index (PI) modified by Tureksy and disclosing images were both recorded. The plaque percent index (PPI) which is defined as the proportion of the dental plaque area and entire teeth surface area. The statistic analysis shows: The Spearman rank correlation coefficient between the PPI of fluorescence image and the PI is 0. 944, and the Pearson correlation coefficient between PPI of fluorescence image and PPI of disclosing image is 0. 875. PPI of fluorescence image has an increasing trend with increase in plaque grade and the PPI show statistically significant differences (p < 0. 01) between different grades. As conclusions, the proposed noninvasive detection system using the optical detecting approach ensures a fairly good accuracy and reliability, besides, the evaluation index of the PPI is more precise compared with the PI We envision it has the potential to be a homecare practice and convince people of the demand for proper oral hygiene.

Dental Plaque Quantification Using Cellular Neural Network-Based Image Segmentation

Analysis of Dental Plaque by Using Cellular Neural Network-Based Image Segmentation

Quantification of Dental Plaque Based on Auto-Fluorescence Imaging

Automatic Dental Plaque Segmentation Based on Local-to-Global Features Fused Self-Attention Network.

DeepPlaq: Dental plaque indexing based on deep neural networks

Dental Plaque Quantification Using FCM-based Classification in HSI Color Space

Automated classification of dual channel dental imaging of auto-fluorescence and white lightby convolutional neural networks

Recognition of Digital Dental X-ray Images Using a Convolutional Neural Network

[Deep Learning-Based Dental Plaque Detection on Permanent Teeth and the Influenced Factors].

A deep learning approach to automatic gingivitis screening based on classification and localization in RGB photos

Deep Learning-Based Multiclass Instance Segmentation for Dental Lesion Detection

Oral screening of dental calculus, gingivitis and dental caries through segmentation on intraoral photographic images using deep learning

Detection of Dental Apical Lesions Using CNNs on Periapical Radiograph

Automated Dental Image Analysis by Deep Learning on Small Dataset

Segmentation and accurate identification of large carious lesions on high quality x-ray images based on Attentional U-Net model. A proof of concept study

Evaluation of dental plaque by quantitative digital image analysis system

Recognition and Segmentation of Teeth and Mandibular Nerve Canals in Panoramic Dental X-rays by Mask RCNN

Intelligently Quantifying the Entire Irregular Dental Structure

Dental Image Segmentation Based on Multiple Linear Regression Correlates with Normalized RGB Values

Exploring the Role of Convolutional Neural Networks (CNN) in Dental Radiography Segmentation: A Comprehensive Systematic Literature Review

Caries and Restoration Detection Using Bitewing Film Based on Transfer Learning with CNNs