Caries Classification Research Articles

Feasibility of the International Caries Classification and Management System (ICCMS) Protocol in a Hospital-Based Setting in India

Feasibility of the International Caries Classification and Management System (ICCMS) Protocol in a Hospital-Based Setting in India

Caries Detection and Classification in Photographs Using an Artificial Intelligence-Based Model-An External Validation Study.

This ex vivo diagnostic study aimed to externally validate a freely accessible AI-based model for caries detection, classification, localisation and segmentation using an independent image dataset. It was hypothesised that there would be no difference in diagnostic performance compared to previously published internal validation data. For the independent dataset, 718 dental images representing different stages of carious (n = 535) and noncarious teeth (n = 183) were retrieved from the internet. All photographs were evaluated by the dental team (reference standard) and the AI-based model (test method). Diagnostic performance was statistically determined using cross-tabulations to calculate accuracy (ACC), sensitivity (SE), specificity (SP) and area under the curve (AUC). An overall ACC of 92.0% was achieved for caries detection, with an ACC of 85.5-95.6%, SE of 42.9-93.3%, SP of 82.1-99.4% and AUC of 0.702-0.909 for the classification of caries. Furthermore, 97.0% of the cases were accurately localised. Fully and partially correct segmentation was achieved in 52.9% and 44.1% of the cases, respectively. The validated AI-based model showed promising diagnostic performance in detecting and classifying caries using an independent image dataset. Future studies are needed to investigate the validity, reliability and practicability of AI-based models using dental photographs from different image sources and/or patient groups.

Methods used for caries detection and diagnosis in Ontario dental practices: a cross-sectional survey

BackgroundEarly detection of caries is essential for applying non-surgical treatment procedures and preventing the formation of cavitated lesions leading to unnecessary removal of tooth structure. Understanding dentists’ preferences for caries detection tools can inform stakeholders about their strategies and knowledge of contemporary, evidence-based caries management approaches. However, there is a lack of research exploring the detection methods of caries commonly used by dentists in Ontario, Canada. The objective of this study was to investigate the methods of caries detection and diagnosis preferred by dentists in Ontario.MethodsA 21-item self-reported survey was mailed to one thousand Ontario dental practices in the Winter of 2022. Descriptive and bivariate data analysis were performed to determine the associations between: demographics and professional practice characteristics (explanatory variables), and methods for detecting and diagnosing dental caries (outcome variables) using SPSS Statistics 29.0.ResultsA total of 325 dentists (33%) responded to the survey, with 274 answering all of the questions completely. The highest proportion of respondents were 35–44 years of age (32.8%) and male (53.4%). More than half of the respondents reported using a dental explorer to assess primary occlusal caries (57.6%), secondary caries (57.1%), and cervical caries (57.5%). Likewise, 57.9% of the participants reported using dental radiographs to diagnose proximal caries. Among additional caries detection tools, digital radiography (89.8%) and traditional radiography (84.7%) were the most used methods/modalities, while cone beam computed tomography was the least (12.8%). Most study participants did not use any caries classification system (77.7%) or caries risk assessment tool (85.3%).ConclusionsParticipants preferred conventional methods for caries detection, instead of contemporary visual-tactile caries lesions classification and/or caries risk assessment systems. These findings indicate a need for continuing dental education programs tailored to evidence-based caries management approaches.

Performance Evaluation of Deep Learning Models for Dental Caries Classification via Panoramic Radiograph Images

Objective: The purpose of this study is to evaluate the ability of deep learning models to classify mandibular molar teeth according to the presence and proximity of caries to the dental pulp. This research summarizes the progress of artificial intelligence and potential dental problems in diagnosis, treatment, and disease prediction in medicine. It discusses data limitations, computational power, ethical considerations, and their implications for dentists. This can lay the groundwork for future research in this rapidly expanding field. Methods: The dataset used in this study consists of 1200 panoramic radiographs, which have been evaluated and classified into three categories: free of dental caries, coded as (H); enamel-dentin caries lesions treated with restorative filling, coded as (R); and deep dental caries that underwent root canal treatment, coded as (E). The images are prepared for the training-testing process using the k-fold crossevaluation technique and then fed into the pre-trained deep learning models for classification. Results: The VGG-19 model achieved superior results compared to the other models, with macro-average scores of 0.9111 for precision, 0.9127 for recall, and 0.9115 for f1-score, respectively. Conclusion: The promising results obtained in this study give confidence in endorsing the use of deep learning models in the dental treatments sector.

Comparison of Two Clinical Approaches FDI Vs Cars Visual Criteria for Secondary Caries Evaluation in Permanent Posterior Teeth

Background: Secondary caries arise at the contact of a natural tooth and a prosthetic repair. The present investigation employed two clinical methodologies, namely Caries associated with restoration or sealants (CARS) and the International Dental Federation (FDI), to evaluate and manage secondary caries in permanent posterior teeth. Objective: The goal of this research is to compare the diagnostic efficacy of FDI and CARS visual criteria for assessing secondary caries in permanent posterior teeth, to identify the more reliable and effective clinical approach for improved treatment planning and patient outcomes. Method: This was conducted at the School of Dentistry Pakistan institute of medical sciences Islamabad, after approval from the ethical committee. The sample size is 160 patients. In this study, two clinical criteria being assessed included one representing the FDI system which includes the marginal staining, marginal adaptation, and caries’ recurrence, while another one was Caries associated with restoration or sealants (CARS) based on ICCMS (International Caries Classification and Management System) for evaluation and treatment of secondary caries. Results: A total 650 restorations were assessed, with a mean age of 30.83 years (SD±2.68). While comparing, there was a strong correlation to the presence of caries lesions (Rho=0.64), and for Marginal adaptation, it was (Rho= -0.45), which depicts a weak inverse correlation. There was a moderate correlation between the two criteria (Rho=0.64,) The majority of CARS visual criteria decisions recommended no treatment in comparison to FDI criteria Conclusion: It is concluded that the FDI system is a more aggressive approach, suggesting a higher number of restoration replacements than the CARS which is less invasive. As a result, the method used to assess secondary caries may result in more or less invasive suggestions for treatment.

Developing the Benchmark: Establishing a Gold Standard for the Evaluation of AI Caries Diagnostics.

Background/Objectives: The aim of this study was to establish a histology-based gold standard for the evaluation of artificial intelligence (AI)-based caries detection systems on proximal surfaces in bitewing images. Methods: Extracted human teeth were used to simulate intraoral situations, including caries-free teeth, teeth with artificially created defects and teeth with natural proximal caries. All 153 simulations were radiographed from seven angles, resulting in 1071 in vitro bitewing images. Histological examination of the carious lesion depth was performed twice by an expert. A total of thirty examiners analyzed all the radiographs for caries. Results: We generated in vitro bitewing images to evaluate the performance of AI-based carious lesion detection against a histological gold standard. All examiners achieved a sensitivity of 0.565, a Matthews correlation coefficient (MCC) of 0.578 and an area under the curve (AUC) of 76.1. The histology receiver operating characteristic (ROC) curve significantly outperformed the examiners' ROC curve (p < 0.001). All examiners distinguished induced defects from true caries in 54.6% of cases and correctly classified 99.8% of all teeth. Expert caries classification of the histological images showed a high level of agreement (intraclass correlation coefficient (ICC) = 0.993). Examiner performance varied with caries depth (p ≤ 0.008), except between E2 and E1 lesions (p = 1), while central beam eccentricity, gender, occupation and experience had no significant influence (all p ≥ 0.411). Conclusions: This study successfully established an unbiased dataset to evaluate AI-based caries detection on bitewing surfaces and compare it to human judgement, providing a standardized assessment for fair comparison between AI technologies and helping dental professionals to select reliable diagnostic tools.

The professional perception of the International Caries Classification and Management System (ICCMS): a pragmatic randomised clinical trial.

Introduction The International Caries Classification and Management System (ICCMS) was developed to standardise caries diagnosis, risk assessment and treatment decisions based on overall evidence. To evaluate its criteria in practice, a clinical trial assessed professionals' perceptions of two caries management systems.Methods A perception questionnaire was administered to two groups: one using ICCMS criteria and the other based on professional experience criteria (non-ICCMS group). The online questionnaire included quantitative scale questions to measure effort and satisfaction and two open-ended questions to collect the positive and negative perceptions of dentists by using either criteria system. The questionnaires were administered six months after the study's implementation.Results Both groups showed high levels of effort and satisfaction. Professionals using ICCMS reported more positive perceptions by citing improved diagnosis (71.4%) and standardised patient care (43%). However, they also noted negative aspects, such as increased number of questionnaires and records (58%), longer clinical sessions (43%) and higher rates of patient absenteeism (29%).Conclusion Professionals who used the ICCMS had a positive outlook on the system, with benefits seen in diagnosis and patient monitoring. However, there is room for improvement in terms of automation and simplification to enhance the professional use of the system in the clinical setting.

Classification of Caries Based on CBCT: A Deep Learning Network Interpretability Study.

This study aimed to create a caries classification scheme based on cone-beam computed tomography (CBCT) and develop two deep learning models to improve caries classification accuracy. A total of 2713 axial slices were obtained from CBCT images of 204 carious teeth. Both classification models were trained and tested using the same pretrained classification networks on the dataset, including ResNet50_vd, MobileNetV3_large_ssld, and ResNet50_vd_ssld. The first model was used directly to classify the original images (direct classification model). The second model incorporated a presegmentation step for interpretation (interpretable classification model). Performance evaluation metrics including accuracy, precision, recall, and F1 score were calculated. The Local Interpretable Model-agnostic Explanations (LIME) method was employed to elucidate the decision-making process of the two models. In addition, a minimum distance between caries and pulp was introduced for determining the treatment strategies for type II carious teeth. The direct model that utilized the ResNet50_vd_ssld network achieved top accuracy, precision, recall, and F1 score of 0.700, 0.786, 0.606, and 0.616, respectively. Conversely, the interpretable model consistently yielded metrics surpassing 0.917, irrespective of the network employed. The LIME algorithm confirmed the interpretability of the classification models by identifying key image features for caries classification. Evaluation of treatment strategies for type II carious teeth revealed a significant negative correlation (p < 0.01) with the minimum distance. These results demonstrated that the CBCT-based caries classification scheme and the two classification models appeared to be acceptable tools for the diagnosis and categorization of dental caries.

Combining public datasets for automated tooth assessment in panoramic radiographs

ObjectivePanoramic radiographs (PRs) provide a comprehensive view of the oral and maxillofacial region and are used routinely to assess dental and osseous pathologies. Artificial intelligence (AI) can be used to improve the diagnostic accuracy of PRs compared to bitewings and periapical radiographs. This study aimed to evaluate the advantages and challenges of using publicly available datasets in dental AI research, focusing on solving the novel task of predicting tooth segmentations, FDI numbers, and tooth diagnoses, simultaneously.Materials and methodsDatasets from the OdontoAI platform (tooth instance segmentations) and the DENTEX challenge (tooth bounding boxes with associated diagnoses) were combined to develop a two-stage AI model. The first stage implemented tooth instance segmentation with FDI numbering and extracted regions of interest around each tooth segmentation, whereafter the second stage implemented multi-label classification to detect dental caries, impacted teeth, and periapical lesions in PRs. The performance of the automated tooth segmentation algorithm was evaluated using a free-response receiver-operating-characteristics (FROC) curve and mean average precision (mAP) metrics. The diagnostic accuracy of detection and classification of dental pathology was evaluated with ROC curves and F1 and AUC metrics.ResultsThe two-stage AI model achieved high accuracy in tooth segmentations with a FROC score of 0.988 and a mAP of 0.848. High accuracy was also achieved in the diagnostic classification of impacted teeth (F1 = 0.901, AUC = 0.996), whereas moderate accuracy was achieved in the diagnostic classification of deep caries (F1 = 0.683, AUC = 0.960), early caries (F1 = 0.662, AUC = 0.881), and periapical lesions (F1 = 0.603, AUC = 0.974). The model’s performance correlated positively with the quality of annotations in the used public datasets. Selected samples from the DENTEX dataset revealed cases of missing (false-negative) and incorrect (false-positive) diagnoses, which negatively influenced the performance of the AI model.ConclusionsThe use and pooling of public datasets in dental AI research can significantly accelerate the development of new AI models and enable fast exploration of novel tasks. However, standardized quality assurance is essential before using the datasets to ensure reliable outcomes and limit potential biases.

Deep learning to support the detection and classification of teeth, dental caries and restorations: a review

Oral health serves as a crucial indicator of individuals' overall well-being and quality of life, making it a pertinent concern regularly addressed by healthcare professionals. Utilizing imaging exams is imperative for detecting and identifying oral diseases and conditions, and the application of Artificial Intelligence (AI) has garnered attention for its potential in this realm. We conducted a systematic literature review focusing on the utilization of Deep Learning techniques in dental radiographs for the detection, segmentation, and classification of teeth, caries, and restorations. Our review encompassed automated searches across prominent databases including the ACM Digital Library, IEEE Xplore Digital Library, PubMed, and Scopus, yielding 393 primary papers published between 2012 and 2023. Following stringent inclusion and exclusion criteria, we thoroughly examined 68 papers, assessing their consistency and adequacy of different aspects in terms of the databases used, techniques implemented, and outcomes reported. It was noted that 41.66% of the analyzed papers lacked clear information regarding approval of data usage from ethics committees. Additionally, despite the interdisciplinary nature of computational techniques in oral health, 38.23% of the surveyed studies were conducted by teams comprising solely professionals from one specific area. Moreover, 66.18% of the papers focused solely on panoramic radiographs, with commonly utilized metrics including accuracy, recall, and precision. Notably, the U-Net and Mask R-CNN networks emerged as the most frequently applied methodologies. Despite the proliferation of investigations in this field, several challenges persist, including the limited availability of public datasets, inadequate detailing of developed methodologies, and a lack of systematization in result presentation. These challenges hinder a fair comparison between studies, presenting a significant obstacle to be addressed for further progress in the field.

Artificial Intelligence for Caries and Tooth Detection in Dental Imaging: A Review

This review delves into the application of artificial intelligence (AI) and deep learning, particularly leveraging convolutional neural networks (CNNs), to enhance dental diagnostics and treatment planning. The primary focus is on the detection and classification of caries as well as the identification of teeth in diverse dental images. A thorough exploration was conducted across databases, including PubMed, IEEE Xplore, and arXiv.org, leading to the identification of 29 pertinent studies. These studies employ various neural network models, encompassing different dental image types and employing diverse performance metrics. The review succinctly outlines the key characteristics and outcomes of these studies, underscoring the remarkable accuracy and the promising potential of AI-driven approaches in the realms of caries detection and tooth identification.&#x0D; Acknowledging the existing limitations within the current body of research, such as small or non-representative datasets, variations in imaging techniques, and a lack of interpretability in deep learning models, the review emphasizes the need for future investigations. It suggests potential research directions aimed at overcoming these challenges, thereby facilitating the seamless integration of AI into routine dental practices.

Tooth caries classification with quantitative light-induced fluorescence (QLF) images using convolutional neural network for permanent teeth in vivo

BackgroundOwing to the remarkable advancements of artificial intelligence (AI) applications, AI-based detection of dental caries is continuously improving. We evaluated the efficacy of the detection of dental caries with quantitative light-induced fluorescence (QLF) images using a convolutional neural network (CNN) model.MethodsOverall, 2814 QLF intraoral images were obtained from 606 participants at a dental clinic using Qraypen C® (QC, AIOBIO, Seoul, Republic of Korea) from October 2020 to October 2022. These images included all the types of permanent teeth of which surfaces were smooth or occlusal. Dataset were randomly assigned to the training (56.0%), validation (14.0%), and test (30.0%) subsets of the dataset for caries classification. Moreover, masked images for teeth area were manually prepared to evaluate the segmentation efficacy. To compare diagnostic performance for caries classification according to the types of teeth, the dataset was further classified into the premolar (1,143 images) and molar (1,441 images) groups. As the CNN model, Xception was applied.ResultsUsing the original QLF images, the performance of the classification algorithm was relatively good showing 83.2% of accuracy, 85.6% of precision, and 86.9% of sensitivity. After applying the segmentation process for the tooth area, all the performance indics including 85.6% of accuracy, 88.9% of precision, and 86.9% of sensitivity were improved. However, the performance indices of each type of teeth (both premolar and molar) were similar to those for all teeth.ConclusionThe application of AI to QLF images for caries classification demonstrated a good performance regardless of teeth type among posterior teeth. Additionally, tooth area segmentation through background elimination from QLF images exhibited a better performance.

Efficacy of Different Fluoride Therapies on Hypersensitive Carious Lesions in Primary Teeth.

Background and Objectives: This prospective, comparative, double-cohort study aimed to compare the efficacy of silver diamine fluoride and potassium iodide (38% SDF+KI; Riva Star®) with sodium fluoride varnish (5% NaF; Duraphat®) in hypersensitive carious lesions in primary teeth to evaluate caries arrest and hypersensitivity relief. Materials and Methods: This study included thirty 2-5-year-olds (mean age = 3.67 ± 1.06 years; 16 males and 14 females) who required a desensitizing treatment for hypersensitive carious defects with visible dentin. A total of 15 of the participants were consecutively allocated to treatment with 5% NaF, and they were further compared to an equal number of participants treated with 38% SDF+KI solutions (n = 15). The treatments were performed following clinical evaluation of caries activity using the International Caries Classification and Management System (ICCMSTM) and the Bjørndal criteria (score of 0-9). Parental-reported hypersensitivity was evaluated using a visual analogue scale (0-10 = no to severe pain). Results: Clinical variables were evaluated at baseline and three months after treatment. Thereafter, a significant decline in hypersensitivity/pain led to lower final scores in the Riva Star® group (0.40 ± 1.12, p = 0.002) than in the Duraphat® group (1.40 ± 2.20, p = 0.004). The caries arrest effect was significantly higher in the Riva Star® group (86.7%) compared to the Duraphat® group three months after treatment (13.3%, p < 0.001). In both groups, there were no statistically significant differences in the children's behavior before, during, and after treatment. Conclusions: Ultimately, with both fluoride therapies reducing hypersensitivity/pain significantly, treatment with 38% SDF+KI was clearly more effective in caries arrest than 5% NaF varnish after a 3-month period.

Microbial Relationship of Carious Deciduous Molars and Adjacent First Permanent Molars.

(1) Epidemiological studies have shown that deciduous molar caries are related to and more severe than permanent molar caries. This study aimed to investigate whether caries subtypes in deciduous molars were associated with caries in first permanent molars and to explore taxonomic and functional profiles of the microbiota involved in different subtypes. (2) 42 mixed-dentition children were recruited and were divided into DMC (carious deciduous molars but caries-free first permanent molars; n = 14), C (carious deciduous and first permanent molars; n = 13), and control (n = 15) groups. Metagenomic sequencing was performed for supragingival plaque samples obtained separately from deciduous and first permanent molars. (3) The microbiota of deciduous molars in the DMC and C groups differed not only in species-based beta diversity but also in compositional and functional profiles. In the C group-like subtype, 14 caries-related species and potential pathways were identified that could be responsible for the caries relationship between the deciduous and permanent molars. In the DMC group-like subtype, the overall functional structure, the levels of Leptotrichia wadei, Streptococcus anginosus, and Stomatobaculum longum and KOs in sugar transporters and fermentation, quorum sensing, and TCA cycle in their first permanent molars surprisingly resembled those of the C group rather than the control group. This suggested that these clinically sound first permanent molars were at a greater risk for caries. (4) Classification of deciduous molar caries according to the microbiota could serve as a caries risk predictor for adjacent first permanent molars.

Microcomputed Tomography Mineral Density Profile as Reference Standard for Early Carious Lesion Activity Assessment

Early caries diagnosis is crucial to treatment decisions in dentistry and requires identification of lesion activity: whether a carious lesion is active (progressively demineralizing) or arrested (progressively remineralizing). This study aimed to identify microtomographic (micro-CT) differences between active and arrested smooth surface enamel lesions, to quantify those micro-CT differences by creating thresholds for ex vivo caries activity assessment to serve as a future reference standard, and to validate those thresholds against the remaining sample. Extracted human permanent teeth (n = 59) were selected for sound surfaces and non-cavitated smooth surface carious lesions. Each surface was then examined for caries activity by calibrated individuals via visual-tactile examination using the International Caries Classification and Management System (ICCMS) activity criteria. Each tooth was scanned via micro-CT and the mineral density was plotted against lesion depth. The area under the curve (AUC) was calculated and represented the loss of density for the outermost 96 μm of enamel. AUC thresholds obtained from micro-CT were established to classify sound, remineralized, and demineralized surfaces against the gold standard examiner’s lesion assessment of sound, inactive, and active lesions, respectively. The established AUC thresholds demonstrated moderate agreement with the assessment in identifying demineralized lesions (k = 0.45), with high sensitivity (0.73) and specificity (0.77). This study demonstrated quantifiable differences among demineralized lesions, remineralized lesions, and sound surfaces, which contributes to the establishment of micro-CT as a reference standard for caries activity that may be used to improve clinical and laboratorial dental caries evaluations.

Artificial intelligence in the detection and classification of dental caries

Automated detection of dental caries could enhance early detection, save clinician time, and enrich treatment decisions. However, a reliable system is lacking. The purpose of this study was to train a deep learning model and to assess its ability to detect and classify dental caries. Bitewings radiographs with a 1876×1402-pixel resolution were collected, segmented, and anonymized with a radiographic image analysis software program and were identified and classified according to the modified King Abdulaziz University (KAU) classification for dental caries. The method was based on supervised learning algorithms trained on semantic segmentation tasks. The mean score for the intersection-over-union of the model was 0.55 for proximal carious lesions on a 5-category segmentation assignment and a mean F1 score of 0.535 using 554 training samples. The study validated the high potential for developing an accurate caries detection model that will expedite caries identification, assess clinician decision-making, and improve the quality of patient care.

Impact of Selective median filter on dental caries classification system using deep learning models

&lt;p&gt;Accurate classification of dental caries is crucial for effective oral healthcare. Filters help to increase exposure of the picture taken for the investigation without degrading image quality. Selective median filter is the chosen preprocessing technique that helps to reduce the noise present in the captured image. Dental caries classification system is a model used to detect the presence of cavity in the given input image. Dental caries classification system is evolved with the use of conventional techniques to artificial neural network. Deep learning models are the artificial neural network models that can able to learn the features from the raw images available in the dataset. If this raw image has noise, then it severely affects the accuracy of the deep learning models. In this paper, impact of the preprocessing technique on the classification accuracy is analyzed. Initially, raw images are taken for training on deep learning models without applying any preprocessing technique. This study investigates the impact of Selective median filtering on a dental caries classification system using deep learning models. The motivation behind this research is to enhance the accuracy and reliability of dental caries diagnosis by reducing noise, removing artifacts, and preserving important details in dental radiographs. Experimental results demonstrate that the implementation of Selective median filtering significantly improves the performance of the deep learning model. The hybrid neural network (HNN) classifier achieves an accuracy of 96.15% with Selective median filtering, outperforming the accuracy of 85.07% without preprocessing. The study highlights the theoretical contribution of Selective median filtering in enhancing dental caries classification systems and emphasizes the practical implications for dental clinics, offering improved diagnostic capabilities and better patient outcomes.&lt;/p&gt;

Gambaran Karies dan Evaluasi Perawatan pada Siswa di Kawasan Puskesmas Ikur Koto, Padang, Sumatera Barat

Caries are still a significant problem in oral and dental health worldwide, including Indonesia. Various efforts must continue to be made to reduce the incidence of caries, both promotive and preventive, as well as curative and rehabilitative. Based on the Riset Kesehatan Dasar 2018, 36.71% of the people of Padang City have cavities/injured/toothache. Of the youth aged 15-24, 37.45% had cavities/injured/toothache. This program aims to get an overview of caries classification and provide appropriate treatment. The overview of caries classification can become data on the pattern of caries development so that strategies for dealing with earlier caries can be carried out appropriately, effectively, and efficiently. The implementation of the present program was carried out through a thorough and comprehensive examination of the oral cavity; then, treatment was carried out according to the complaints and cases found. The post-treatment evaluation was carried out by distributing questionnaires to participants afterward. Based on the depth of caries treated, 57.14% were dentin caries. Caries classification based on G.V. Black showed that Class I cavity was the highest incidence (64.29%). Mount and Hume's classification presented class 1.3 (39.28%) as the highest, while based on the ICDAS classification, it was class D3 (46.44%). The post-treatment evaluation showed that all participants (100%) stated that the condition of the teeth improved and that there were no complaints/discomfort after treatment.

Enhanced Clinical Decision-Making and Delivery of Minimally Invasive Care Using the ICCMS4D Integrated with Hands-Free Fluorescence-Based Loupes and a Chemomechanical Caries Removal Agent.

The purpose of this case report was to evaluate an approach for enhanced clinical decision-making while providing minimally invasive treatment, using the combined International Caries Classification and Management System (ICCMS 4D) with hands-free fluorescence-enhanced loupes (Reveal, Designs for Vision Inc., New York City, NY, United States) and a chemo-mechanical caries removal agent (Papacárie Duo Gel, Formula & Acao, Sao Paulo, Brazil). In recent decades, a shift towards a conservative approach to caries management has developed. The use of adequate operative techniques and correct decision-making are of paramount importance for early caries detection and tooth-preserving operative care. The use of noninvasive fluorescence-based adjuncts for enhanced diagnostic accuracy has gained popularity. Fluorescence describes the absorption of photons by a molecule, followed by its partial emission at a longer wavelength. Fluorescence detection of bacterial activity is largely based on the red/orange fluorescence emanating from bacterial metabolism, and has been shown to be found particularly in active caries and mature anaerobic plaque. The combined approach of using the standardized visual International Caries Detection and Assessment System (ICDAS) with fluorescence as an operative adjunct was shown to enhance the advantages of both systems. The approach may potentially increase detection sensitivity without compromising the specificity of the visual tactile method. A recent hands-free device is aimed to combine simultaneous diagnosis and therapy (theragnosis) using fluorescence, while overcoming possible obstacles to workflow continuity. The "Reveal" fluorescence-enhanced loupes are fitted with a dual white light/fluorescence light, and are said to allow the clinician to conduct any operational procedure with optional fluorescence guidance, from detection to caries removal, to treatment completion. Using the described method, a comprehensive pathway that emphasizes evidence-based information, correct clinical decision-making, and a minimally invasive therapeutic approach was proposed. The approach may represent a potential advancement in providing minimally invasive caries care.

One-Stage Methods of Computer Vision Object Detection to Classify Carious Lesions from Smartphone Imaging

The current study aimed to implement and validate an automation system to detect carious lesions from smartphone images using different one-stage deep learning techniques. 233 images of carious lesions were captured using a smartphone camera system at 1432 × 1375 pixels, then classified and screened according to a visual caries classification index. Following data augmentation, the YOLO v5 model for object detection was used. After training the model with 1452 images at 640 × 588 pixel resolution, which included the ones that were created via image augmentation, a discrimination experiment was performed. Diagnostic indicators such as true positive, true negative, false positive, false negative, and mean average precision were used to analyze object detection performance and segmentation of systems. YOLO v5X and YOLO v5M models achieved superior performance over the other models on the same dataset. YOLO v5X’s mAP was 0.727, precision was 0.731, and recall was 0.729, which was higher than other models of YOLO v5, which generated 64% accuracy, with YOLO v5M producing slightly inferior results. Overall mAPs of 0.70, precision of 0.712, and recall of 0.708 were achieved. Object detection through the current YOLO models was able to successfully extract and classify regions of carious lesions from smartphone photographs of in vitro tooth specimens with reasonable accuracy. YOLO v5M was better fit to detect carious microcavitations while YOLO v5X was able to detect carious changes without cavitation. No single model was capable of adequately diagnosing all classifications of carious lesions.