Deep-learning approach for caries detection and segmentation on dental bitewing radiographs

Bayrakdar, İ̇brahim Şevki; Orhan, Kaan; Akarsu, Serdar; Çelik, Özer; Atasoy, Samet; Pekince, Adem; Yaşa, Yasin; Bilgir, Elif; Sağlam, Hande; Aslan, Ahmet Faruk; Odabaş, Alper

doi:10.1007/s11282-021-00577-9

Cited by 65 publications

(49 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This could be cost- effective, due to its higher sensitivity in the detection and arrest of early caries lesions [ 112 , 113 ]. CNN-based AI algorithms can be beneficial to the dentist as clinical decision support systems [ 114 ]. A deep CNN system can be used to number teeth in bitewing radiographs and save the dentist time by automatically preparing dental charts [ 115 ].…”

Section: Resultsmentioning

confidence: 99%

Where Is the Artificial Intelligence Applied in Dentistry? Systematic Review and Literature Analysis

et al. 2022

View full text Add to dashboard Cite

This literature research had two main objectives. The first objective was to quantify how frequently artificial intelligence (AI) was utilized in dental literature from 2011 until 2021. The second objective was to distinguish the focus of such publications; in particular, dental field and topic. The main inclusion criterium was an original article or review in English focused on dental utilization of AI. All other types of publications or non-dental or non-AI-focused were excluded. The information sources were Web of Science, PubMed, Scopus, and Google Scholar, queried on 19 April 2022. The search string was “artificial intelligence” AND (dental OR dentistry OR tooth OR teeth OR dentofacial OR maxillofacial OR orofacial OR orthodontics OR endodontics OR periodontics OR prosthodontics). Following the removal of duplicates, all remaining publications were returned by searches and were screened by three independent operators to minimize the risk of bias. The analysis of 2011–2021 publications identified 4413 records, from which 1497 were finally selected and calculated according to the year of publication. The results confirmed a historically unprecedented boom in AI dental publications, with an average increase of 21.6% per year over the last decade and a 34.9% increase per year over the last 5 years. In the achievement of the second objective, qualitative assessment of dental AI publications since 2021 identified 1717 records, with 497 papers finally selected. The results of this assessment indicated the relative proportions of focal topics, as follows: radiology 26.36%, orthodontics 18.31%, general scope 17.10%, restorative 12.09%, surgery 11.87% and education 5.63%. The review confirms that the current use of artificial intelligence in dentistry is concentrated mainly around the evaluation of digital diagnostic methods, especially radiology; however, its implementation is expected to gradually penetrate all parts of the profession.

show abstract

Section: Resultsmentioning

confidence: 99%

Where Is the Artificial Intelligence Applied in Dentistry? Systematic Review and Literature Analysis

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Trials were also made with the object detection method using Yolov5 architecture in order to increase the success rates in tooth numbering. In the statistical evaluation, the performance of the system was evaluated by using the confusion matrix system [20] and receiver operating characteristic (ROC) analysis. Sensitivity, precision, F1 score, accurancy were calculated with the confusion matrix system.…”

Section: Materials Methodsmentioning

confidence: 99%

“…When the AI-based studies in eld of dentistry are examined, it is seen that there are more studies on dental radiographs. In these studies, it has been tried to detect pathologies such as caries [20], apical lesion [21], dental restoration [22] and periodontal disease [23,24] from different types of radiographs by using AI systems [25]. Despite this, it is seen that the number of studies aiming to make diagnosis and treatment planning through dental photographs using AI algorithms and evaluating the usability of these systems as assistive systems for physicians is quite limited [22,[26][27][28][29][30].…”

Section: Introductionmentioning

confidence: 99%

Detection of Tooth Numbering, Frenulum, Gingival Hyperplasia and Gingival Inflammation on Dental Photographs Using Convolutional Neural Network Algorithms: An Initial Study

Bayrakdar

Uğurlu

Yavuz

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Objectives The aim of this study is to perform tooth numbering using deep learning algorithms on digital dental photographs, and to evaluate the success of these algorithms in determining the presence of frenulum, gingival hyperplasia and gingival inflammation which play an important role in periodontal treatment planning. Materials and Methods Six-hundred-fifty-four (n = 654) intraoral photographs were included in the study. A total of 16795 teeth in all photographs were segmented and the numbering of the teeth was carried out according to the FDI system. Two-thousand-four-hundred-and-ninety-three frenulum attachments (n = 2493), 1211 gingival hyperplasia areas and 2956 gingival inflammation areas in the photographs were labeled using the segmentation method. Images were sized before artificial intelligence (AI) training and data set was separated as training, validation and test groups. Yolov5 architecture were used in the creation of the models. The confusion matrix system and ROC analysis were used in the statistical evaluation of the results. Results When results of study were evaluated; sensitivity, precision, F1 score and AUC for tooth numbering were 0.990, 0.784, 0.875, 0.989; for frenulum attachments were 0.894, 0.775, 0.830 and 0.827; for gingival hyperplasia were 0.757, 0.675, 0.714, 0.774; for gingival inflammation were 0.737, 0.823, 0.777, 0.802 (respectively). Conclusions There is a need for more comprehensive studies to be carried out on this subject by increasing the number of data and the number of parameters evaluated. Clinical relevance The current study showed that in the future, periodontal problem determination from dental photographs could be performed using AI systems.

show abstract

“…They have been applied to the detection of lung cancer from chest X-ray images 8 , determination of retinal detachment 9 , detection of osteoporosis 10 , screening of breast cancer 11 , etc. In addition, many deep learning-related studies have been reported in the field of dentistry, and classifiers have been developed for areas such as caries 12 , periapical lesions 13 , dental implants 14 , maxillary sinusitis 15 , and position classification of the mandibular third molars 16 . Furthermore, deep learning has occasionally been more accurate than human diagnosis 17 , 18 .…”

Section: Introductionmentioning

confidence: 99%

Deep learning model for analyzing the relationship between mandibular third molar and inferior alveolar nerve in panoramic radiography

Sukegawa

Tanaka

Hara

et al. 2022

Sci Rep

View full text Add to dashboard Cite

In this study, the accuracy of the positional relationship of the contact between the inferior alveolar canal and mandibular third molar was evaluated using deep learning. In contact analysis, we investigated the diagnostic performance of the presence or absence of contact between the mandibular third molar and inferior alveolar canal. We also evaluated the diagnostic performance of bone continuity diagnosed based on computed tomography as a continuity analysis. A dataset of 1279 images of mandibular third molars from digital radiographs taken at the Department of Oral and Maxillofacial Surgery at a general hospital (2014–2021) was used for the validation. The deep learning models were ResNet50 and ResNet50v2, with stochastic gradient descent and sharpness-aware minimization (SAM) as optimizers. The performance metrics were accuracy, precision, recall, specificity, F1 score, and area under the receiver operating characteristic curve (AUC). The results indicated that ResNet50v2 using SAM performed excellently in the contact and continuity analyses. The accuracy and AUC were 0.860 and 0.890 for the contact analyses and 0.766 and 0.843 for the continuity analyses. In the contact analysis, SAM and the deep learning model performed effectively. However, in the continuity analysis, none of the deep learning models demonstrated significant classification performance.

show abstract

Deep-learning approach for caries detection and segmentation on dental bitewing radiographs

Cited by 65 publications

References 45 publications

Where Is the Artificial Intelligence Applied in Dentistry? Systematic Review and Literature Analysis

Where Is the Artificial Intelligence Applied in Dentistry? Systematic Review and Literature Analysis

Detection of Tooth Numbering, Frenulum, Gingival Hyperplasia and Gingival Inflammation on Dental Photographs Using Convolutional Neural Network Algorithms: An Initial Study

Deep learning model for analyzing the relationship between mandibular third molar and inferior alveolar nerve in panoramic radiography

Contact Info

Product

Resources

About