ABSTRACT
INTRODUCTION: Generative artificial intelligence (GenAI) Large-language models such as ChatGPT have become increasingly popular in various fields. However, the impact of ChatGPT on dental research writing has yet to be quantified. This study aimed to assess ChatGPT's usage in dental research writing and discuss potential advantages and challenges. METHODS: Using a bibliometric design, we performed a keyword analysis of specific 'signaling words' indicative of ChatGPT use in the titles/abstracts of 299,695 dental research abstracts indexed PubMed 2018-2024. Statistical comparisons using normalized ratios per 10,000 dental publications compared changes in word frequency before and after the ChatGPT release on November 30, 2022. RESULTS: Before ChatGPT's release, the frequency of abstracts with signaling words was 47.1 per 10,000 papers. After the release, this increased to 224.2 per 10,000 papers, an increase of 177.2 per 10,000 papers (p = 0.014, 95 % CI 53.5-300.7). The word 'delve' showed the most significant usage increase (increased ratio=17.0). CONCLUSIONS: This study is among the first to systematically assess the use of GenAI, specifically ChatGPT, in dental research. We found evidence of the use and growth of ChatGPT in dental research publications. This trend indicates the widespread adoption of GenAI-assisted writing in scientific communication, consistent with other scientific fields. While GenAI can potentially increase productivity and inclusivity, it raises concerns such as bias, inaccuracy, and distortion of academic incentives. Therefore, our findings support the need for clear AI guidelines and standards for academic publishing to ensure responsible use and maintain scientific integrity.
ABSTRACT
BACKGROUND: Toothpastes available in Europe contain a range of fluoride concentrations, with some meeting the recommended level for caries prevention (>1000 ppm fluoride (ppm)) and others containing low or no fluoride. This study evaluated toothpaste fluoride concentrations in Latvia and Lithuania to inform targeted public health strategies in regions with a high prevalence of dental caries. METHODS: This cross-sectional study was conducted from May 2019 to May 2020, using a validated questionnaire in Latvia and Lithuania. Nationally representative samples (1309 families and 5436 members) provided data through a mixed-mode survey (paper and online) on sociodemographic information, toothpaste type, brand, and type. Descriptive statistics and chi-square tests (p < 0.05) were used for analysis. RESULTS: Fifteen percent of families used non-fluoridated toothpaste and 12% used <1000 part per million (ppm) fluoride. In Latvia, 56.8% of preschoolers and 28.7% of schoolchildren used <1000 ppm or non-fluoride toothpaste, whereas in Lithuania, 47.2% of preschoolers and 29.1% of schoolchildren used <1000 ppm or non-fluoride toothpaste; 63% of adolescents and 73% of adults used toothpaste with optimal fluoride content (≥1000 ppm). Of the 228 registered toothpaste types, 62% contained more than 1000 ppm, which is optimal for caries prevention; 29% of Latvian and 24% of Lithuanian families used at least one non-fluoridated toothpaste. CONCLUSION: This study revealed significant gaps in the use of fluoride toothpaste among families in Latvia and Lithuania, especially among children. To effectively prevent dental caries, targeted interventions, and education must promote optimal fluoride toothpaste use, particularly among vulnerable populations.
ABSTRACT
Knowledge gaps exist regarding optimal silver diammine fluoride (SDF) regimens and the efficacy of new products for arresting dental caries in young children. We evaluated the effectiveness of 38%-SDF (SDI-RivaStar), Tiefenfluorid (TF) comparing with Placebo (P), all in conjunction with behavioural modification (BM), in preventing major complications (endodontic/extractions/pain)-a patient-centred outcome-due to early childhood caries over 12 months in children under 71-months. A six-arm, patient/parent-blinded, superiority, placebo-controlled randomised control trial at the university clinic in Riga, Latvia, from 1/9/20-31/8/22 (Protocol registration ISRCTN17005348). The trial tested six protocols, using three compounds (P/SDF/TF) under two regimes: annual and biannual (P1/P2/TF1/TF2/SDF1/SDF2) for major complications. Secondary outcomes included minor complications and parental satisfaction. All groups received BM. 373/427 randomised children (87.3%) completed the study. SDF2 had a significantly lower rate and risk of major (21.5%, OR = 0.28, 95%CI [0.11, 0.72], p < 0.05) and minor complications (OR = 0.16 (95%CI [0.05, 0.50], p = 0.002). Overall satisfaction was 96% (p > 0.05). SDF biannual application with BM effectively prevented major complications of early childhood caries and was well accepted by children and their parents. Trial registration number: ISRCTN17005348, principal investigator: Ilze Maldupa, registration date: 30/06/2021.Clinical trial registration number: ISRCTN registry: ISRCTN17005348, https://doi.org/10.1186/ISRCTN17005348 , registration date: 30/06/2021.
Subject(s)
Cariostatic Agents , Dental Caries , Child, Preschool , Humans , Cariostatic Agents/therapeutic use , Dental Caries/prevention & control , Dental Caries/drug therapy , Dental Caries Susceptibility , Fluoride Treatment , Fluorides, Topical/therapeutic use , Parents , Patient Satisfaction , Personal Satisfaction , Quaternary Ammonium Compounds/therapeutic use , Silver Compounds/therapeutic use , Treatment Outcome , InfantABSTRACT
INTRODUCTION: Interest is growing in the potential of artificial intelligence (AI) chatbots and large language models like OpenAI's ChatGPT and Google's Gemini, particularly in dental education. To explore dental educators' perceptions of AI chatbots and large language models, specifically their potential benefits and challenges for dental education. MATERIALS AND METHODS: A global cross-sectional survey was conducted in May-June 2023 using a 31-item online-questionnaire to assess dental educators' perceptions of AI chatbots like ChatGPT and their influence on dental education. Dental educators, representing diverse backgrounds, were asked about their use of AI, its perceived impact, barriers to using chatbots, and the future role of AI in this field. RESULTS: 428 dental educators (survey views = 1516; response rate = 28%) with a median [25/75th percentiles] age of 45 [37, 56] and 16 [8, 25] years of experience participated, with the majority from the Americas (54%), followed by Europe (26%) and Asia (10%). Thirty-one percent of respondents already use AI tools, with 64% recognising their potential in dental education. Perception of AI's potential impact on dental education varied by region, with Africa (4[4-5]), Asia (4[4-5]), and the Americas (4[3-5]) perceiving more potential than Europe (3[3-4]). Educators stated that AI chatbots could enhance knowledge acquisition (74.3%), research (68.5%), and clinical decision-making (63.6%) but expressed concern about AI's potential to reduce human interaction (53.9%). Dental educators' chief concerns centred around the absence of clear guidelines and training for using AI chatbots. CONCLUSION: A positive yet cautious view towards AI chatbot integration in dental curricula is prevalent, underscoring the need for clear implementation guidelines.
ABSTRACT
OBJECTIVE: Authors reported multiple definitions of e-oral health and related terms, and used several definitions interchangeably, like mhealth, teledentistry, teleoral medicine and telehealth. The International Association of Dental Research e-Oral Health Network (e-OHN) aimed to establish a consensus on terminology related to digital technologies used in oral healthcare. METHOD: The Crowdsourcing Delphi method used in this study comprised of four main stages. In the first stage, the task force created a list of terms and definitions around digital health technologies based on the literature and established a panel of experts. Inclusion criteria for the panellists were: to be actively involved in either research and/or working in e-oral health fields; and willing to participate in the consensus process. In the second stage, an email-based consultation was organized with the panel of experts to confirm an initial set of terms. In the third stage, consisted of: a) an online meeting where the list of terms was presented and refined; and b) a presentation at the 2022-IADR annual meeting. The fourth stage consisted of two rounds of feedback to solicit experts' opinion about the terminology and group discussion to reach consensus. A Delphi-questionnaire was sent online to all experts to independently assess a) the appropriateness of the terms, and b) the accompanying definitions, and vote on whether they agreed with them. In a second round, each expert received an individualised questionnaire, which presented the expert's own responses from the first round and the panellists' overall response (% agreement/disagreement) to each term. It was decided that 70% or higher agreement among experts on the terms and definitions would represent consensus. RESULTS: The study led to the identification of an initial set of 43 terms. The list of initial terms was refined to a core set of 37 terms. Initially, 34 experts took part in the consensus process about terms and definitions. From them, 27 experts completed the first rounds of consultations, and 15 the final round of consultations. All terms and definitions were confirmed via online voting (i.e., achieving above the agreed 70% threshold), which indicate their agreed recommendation for use in e-oral health research, dental public health, and clinical practice. CONCLUSION: This is the first study in oral health organised to achieve consensus in e-oral health terminology. This terminology is presented as a resource for interested parties. These terms were also conceptualised to suit with the new healthcare ecosystem and the place of e-oral health within it. The universal use of this terminology to label interventions in future research will increase the homogeneity of future studies including systematic reviews.
Subject(s)
Ecosystem , Oral Health , Humans , ConsensusABSTRACT
BACKGROUND/AIM: This study aimed to develop a protocol that combines cone-beam computed tomography (CBCT), software, and 3D printing to design 3D replicas for tooth autotransplantation. The goal was to evaluate the impact of this approach on the extraoral time of the donor teeth and the total surgical time, thereby enhancing surgical efficiency and outcomes. MATERIALS AND METHODS: A non-randomized trial (protocol 10.1186/ISRCTN13563091) was conducted at Riga Stradins University, enrolling 46 patients (13-22 years old) who required molar extraction and possessed a non-erupted third molar. The patients were sequentially assigned to a 3D replica group (24 patients) or a control group (22 patients). The primary outcome measured was the extra-alveolar time of the donor tooth, and the secondary outcome was the total duration of surgery. Both were assessed using a sample size capable of detecting a 10-min difference. A generalized linear model adjusted for various factors was used to test for significant time differences (p < .05) between the groups. RESULTS: Forty-six patients were included in this analysis. The effect of using 3D replicas was not statistically significant and was associated with a decrease in the extraoral time of the donor tooth in seconds (ß = -9.35, 95% CI [-40.86, 22.16]). For the total surgical time in minutes, the use of 3D replicas had a statistically significant impact, reducing the operation duration in minutes (ß = -13.42, 95% CI [-24.50, -2.34]). No early complications were observed in either group, with all teeth present at 3-4 weeks post-surgery. CONCLUSIONS: The integration of 3D printing technology can enhance the efficiency of autotransplantation surgeries, primarily by reducing surgical time.
Subject(s)
Molar, Third , Molar , Adolescent , Adult , Humans , Young Adult , Cone-Beam Computed Tomography/methods , Molar, Third/surgery , Molar, Third/transplantation , Printing, Three-Dimensional , Software , Transplantation, Autologous/methodsABSTRACT
Background: Child neglect is a public health, human rights, and social problem, with potentially devastating and costly consequences. The aim of this study was to: (1) summarize the oral health profile of children across the globe; (2) provide a brief overview of legal instruments that can offer children protection from dental neglect; and (3) discuss the effectiveness of these legal instruments. Methods: We summarized and highlighted the caries profile and status of implementation of legislation on child dental neglect for 26 countries representing the World Health Organization regions: five countries in Africa (Nigeria, South Africa, Sudan, Tanzania, Zimbabwe), eight in the Americas (Argentina, Brazil, Canada, Chile, Mexico, Peru, Unites States of America, Uruguay), six in the Eastern Mediterranean (Egypt, Iran, Libya, Jordan, Qatar, Saudi Arabia), four in Europe (Italy, Latvia, Serbia, United Kingdom), two in South-East Asia (India and Indonesia) and one country (China) with its special administrative region (Hong Kong) in the Western Pacific. Results: Twenty-five of the 26 countries have legal instruments to address child neglect. Only two (8.0%) of these 25 countries had specific legal instruments on child dental neglect. Although child neglect laws can be interpreted to establish a case of child dental neglect, the latter may be difficult to establish in countries where governments have not addressed barriers that limit children's access to oral healthcare. Where there are specific legal instruments to address child dental neglect, a supportive social ecosystem has also been built to facilitate children's access to oral healthcare. A supportive legal environment, however, does not seem to confer extra protection against risks for untreated dental caries. Conclusions: The institution of specific country-level legislation on child dental neglect may not significantly reduce the national prevalence of untreated caries in children. It, however, increases the prospect for building a social ecosystem that may reduce the risk of untreated caries at the individual level. Social ecosystems to mitigate child dental neglect can be built when there is specific legislation against child dental neglect. It may be more effective to combine public health and human rights-based approaches, inclusive of an efficient criminal justice system to deal with child dental neglect.
ABSTRACT
BACKGROUND: We aimed to assess the adherence to five transparency practices (data availability, code availability, protocol registration and conflicts of interest (COI), and funding disclosures) from open access Coronavirus disease 2019 (COVID-19) related articles. METHODS: We searched and exported all open access COVID-19-related articles from PubMed-indexed journals in the Europe PubMed Central database published from January 2020 to June 9, 2022. With a validated and automated tool, we detected transparent practices of three paper types: research articles, randomized controlled trials (RCTs), and reviews. Basic journal- and article-related information were retrieved from the database. We used R for the descriptive analyses. RESULTS: The total number of articles was 258,678, of which we were able to retrieve full texts of 186,157 (72%) articles from the database Over half of the papers (55.7%, n = 103,732) were research articles, 10.9% (n = 20,229) were review articles, and less than one percent (n = 1,202) were RCTs. Approximately nine-tenths of articles (in all three paper types) had a statement to disclose COI. Funding disclosure (83.9%, confidence interval (CI): 81.7-85.8 95%) and protocol registration (53.5%, 95% CI: 50.7-56.3) were more frequent in RCTs than in reviews or research articles. Reviews shared data (2.5%, 95% CI: 2.3-2.8) and code (0.4%, 95% CI: 0.4-0.5) less frequently than RCTs or research articles. Articles published in 2022 had the highest adherence to all five transparency practices. Most of the reviews (62%) and research articles (58%) adhered to two transparency practices, whereas almost half of the RCTs (47%) adhered to three practices. There were journal- and publisher-related differences in all five practices, and articles that did not adhere to transparency practices were more likely published in lowest impact journals and were less likely cited. CONCLUSION: While most articles were freely available and had a COI disclosure, adherence to other transparent practices was far from acceptable. A much stronger commitment to open science practices, particularly to protocol registration, data and code sharing, is needed from all stakeholders.
Subject(s)
COVID-19 , Humans , COVID-19/epidemiology , Publications , Conflict of Interest , Disclosure , EuropeABSTRACT
Aim: The COVID-19 pandemic has accelerated teledentistry research with great interest reflected in the increasing number of publications. In many countries, teledentistry programs were established although not much is known about the extent of incorporating teledentistry into practice and healthcare systems. This study aimed to report on policies and strategies related to teledentistry practice as well as barriers and facilitators for this implementation in 19 countries. Methods: Data were presented per country about information and communication technology (ICT) infrastructure, income level, policies for health information system (HIS), eHealth and telemedicine. Researchers were selected based on their previous publications in teledentistry and were invited to report on the situation in their respective countries including Bosnia and Herzegovina, Canada, Chile, China, Egypt, Finland, France, Hong Kong SAR, Iran, Italy, Libya, Mexico, New Zealand, Nigeria, Qatar, Saudi Arabia, South Africa, United Kingdom, Zimbabwe. Results: Ten (52.6%) countries were high income, 11 (57.9%) had eHealth policies, 7 (36.8%) had HIS policies and 5 (26.3%) had telehealth policies. Six (31.6%) countries had policies or strategies for teledentistry and no teledentistry programs were reported in two countries. Teledentistry programs were incorporated into the healthcare systems at national (n = 5), intermediate (provincial) (n = 4) and local (n = 8) levels. These programs were established in three countries, piloted in 5 countries and informal in 9 countries. Conclusion: Despite the growth in teledentistry research during the COVID-19 pandemic, the use of teledentistry in daily clinical practice is still limited in most countries. Few countries have instituted teledentistry programs at national level. Laws, funding schemes and training are needed to support the incorporation of teledentistry into healthcare systems to institutionalize the practice of teledentistry. Mapping teledentistry practices in other countries and extending services to under-covered populations increases the benefit of teledentistry.
ABSTRACT
OBJECTIVE: Federated Learning (FL) enables collaborative training of artificial intelligence (AI) models from multiple data sources without directly sharing data. Due to the large amount of sensitive data in dentistry, FL may be particularly relevant for oral and dental research and applications. This study, for the first time, employed FL for a dental task, automated tooth segmentation on panoramic radiographs. METHODS: We employed a dataset of 4,177 panoramic radiographs collected from nine different centers (n = 143 to n = 1881 per center) across the globe and used FL to train a machine learning model for tooth segmentation. FL performance was compared against Local Learning (LL), i.e., training models on isolated data from each center (assuming data sharing not to be an option). Further, the performance gap to Central Learning (CL), i.e., training on centrally pooled data (based on data sharing agreements) was quantified. Generalizability of models was evaluated on a pooled test dataset from all centers. RESULTS: For 8 out of 9 centers, FL outperformed LL with statistical significance (p<0.05); only the center providing the largest amount of data FL did not have such an advantage. For generalizability, FL outperformed LL across all centers. CL surpassed both FL and LL for performance and generalizability. CONCLUSION: If data pooling (for CL) is not feasible, FL is shown to be a useful alternative to train performant and, more importantly, generalizable deep learning models in dentistry, where data protection barriers are high. CLINICAL SIGNIFICANCE: This study proves the validity and utility of FL in the field of dentistry, which encourages researchers to adopt this method to improve the generalizability of dental AI models and ease their transition to the clinical environment.
Subject(s)
Artificial Intelligence , Deep Learning , Humans , Radiography, Panoramic , Research PersonnelABSTRACT
BACKGROUND: Osteoporosis is a disease which is characterized by a decrease in general bone mineral density (BMD), resulting in decreased bone strength and an increased risk of bone fractures. The effect of reduced BMD on the jawbones is still not fully understood. The purpose of the study was to evaluate the impact of BMD on the quality and quantity of the edentulous mandible. METHODS: The present study included 127 edentulous postmenopausal women who underwent cone beam computed tomography (CBCT) examinations. BMD measurements of the lumbar spine and femoral necks were performed by dual-energy X-ray absorptiometry. In the cross-sectional CBCT images, three different areas of the mandible (lateral incisor, first premolar, and first molar) were selected. The complete mandibular, trabecular, and cortical bone volumes were measured. All measurements were performed on the total mandibular area, and the basal and alveolar parts of the mandible. RESULTS: The volume of the cortical bone was reduced for females with reduced BMD in the lateral incisor and first premolar regions, both in the total mandibular area and in the basal part of the mandible. The trabecular bone volume statistically significantly increased when the BMD decreased in the complete mandibular area and the basal part of the mandible (linear regression). The total bone volume significantly decreased with a decrease in BMD in the basal part of the mandible. CONCLUSIONS: Reduced BMD has a negative effect on the quantity and quality of bone in the basal part of the edentulous mandible.
ABSTRACT
OBJECTIVES: Artificial intelligence (AI) is swiftly entering oral health services and dentistry, while most providers show limited knowledge and skills to appraise dental AI applications. We aimed to define a core curriculum for both undergraduate and postgraduate education, establishing a minimum set of outcomes learners should acquire when taught about oral and dental AI. METHODS: Existing curricula and other documents focusing on literacy of medical professionals around AI were screened and relevant items extracted. Items were scoped and adapted using expert interviews with members of the IADR's e-oral health group, the ITU/WHO's Focus Group AI for Health and the Association for Dental Education in Europe. Learning outcome levels were defined and each item assigned to a level. Items were systematized into domains and a curricular structure defined. The resulting curriculum was consented using an online Delphi process. RESULTS: Four domains of learning outcomes emerged, with most outcomes being on the "knowledge" level: (1) Basic definitions and terms, the reasoning behind AI and the principle of machine learning, the idea of training, validating and testing models, the definition of reference tests, the contrast between dynamic and static AI, and the problem of AI being a black box and requiring explainability should be known. (2) Use cases, the required types of AI to address them, and the typical setup of AI software for dental purposes should be taught. (3) Evaluation metrics, their interpretation, the relevant impact of AI on patient or societal health outcomes and associated examples should be considered. (4) Issues around generalizability and representativeness, explainability, autonomy and accountability and the need for governance should be highlighted. CONCLUSION: Both educators and learners should consider this core curriculum during planning, conducting and evaluating oral and dental AI education. CLINICAL SIGNIFICANCE: A core curriculum on oral and dental AI may help to increase oral and dental healthcare providers' literacy around AI, allowing them to critically appraise AI applications and to use them consciously and on an informed basis.
Subject(s)
Artificial Intelligence , Education, Dental , Humans , Curriculum , Delivery of Health Care , Health PersonnelABSTRACT
We assessed adherence to five transparency practices-data sharing, code sharing, conflict of interest disclosure, funding disclosure, and protocol registration-in articles in dental journals. We searched and exported the full text of all research articles from PubMed-indexed dental journals available in the Europe PubMed Central database until the end of 2021. We programmatically assessed their adherence to the five transparency practices using a validated and automated tool. Journal- and article-related information was retrieved from ScimagoJR and Journal Citation Reports. Of all 329,784 articles published in PubMed-indexed dental journals, 10,659 (3.2%) were available to download. Of those, 77% included a conflict of interest disclosure, and 62% included a funding disclosure. Seven percent of the articles had a registered protocol. Data sharing (2.0%) and code sharing (0.1%) were rarer. Sixteen percent of articles did not adhere to any of the five transparency practices, 29% adhered to one, 48% adhered to two, 7.0% adhered to three, 0.3% adhered to four, and no article adhered to all five practices. Adherence to transparency practices increased over time; however, data and code sharing especially remained rare. Coordinated efforts involving all stakeholders are needed to change current transparency practices in dental research.
Subject(s)
Dental Research , Disclosure , EuropeABSTRACT
BACKGROUND: During the COVID19 pandemic, the Latvian government issued first absolute restrictions (elective treatments prohibited, only emergency care) and later relative restrictions (preference for non-aerosol-generating procedures (AGP) and emergency care) on dental care. This study aims to assess the impact of these restrictions on the decision made by Latvian dentists about caries treatment. METHODS: A Survey-based cross-sectional study was used. A minimum sample size of 174 dentists was estimated for national representativeness (N = 1524). The questionnaire was developed by experts and sent three times via email to Latvian dentists from July to September 2020 and was also delivered in printed form at two national conferences in September and October 2020. Descriptive statistics were calculated. RESULTS: We received 373 completed questionnaires, with a total response rate of 24.5%. Under the recommendation to reduce AGP for the treatment of uncomplicated caries, 10% of the dentists stated that they would stop attending, 54% would only attend emergencies, and 36% would attend as usual. Under prohibition, the percentages are 15%, 74%, and 11%, respectively. Regarding the type of treatment, more than 75% would opt to proceed with selective caries removal for both primary and permanent teeth and 10% for extraction. CONCLUSION: Latvian dentists are willing to treat patients with caries during the pandemic and state that they prefer to use non- or minimally invasive and less aerosol-generating methods for caries treatment.
Subject(s)
COVID-19 , Dental Caries , Cross-Sectional Studies , Dental Caries/etiology , Dental Caries/therapy , Dentistry , Dentists , HumansABSTRACT
The detection and classification of cystic lesions of the jaw is of high clinical relevance and represents a topic of interest in medical artificial intelligence research. The human clinical diagnostic reasoning process uses contextual information, including the spatial relation of the detected lesion to other anatomical structures, to establish a preliminary classification. Here, we aimed to emulate clinical diagnostic reasoning step by step by using a combined object detection and image segmentation approach on panoramic radiographs (OPGs). We used a multicenter training dataset of 855 OPGs (all positives) and an evaluation set of 384 OPGs (240 negatives). We further compared our models to an international human control group of ten dental professionals from seven countries. The object detection model achieved an average precision of 0.42 (intersection over union (IoU): 0.50, maximal detections: 100) and an average recall of 0.394 (IoU: 0.50-0.95, maximal detections: 100). The classification model achieved a sensitivity of 0.84 for odontogenic cysts and 0.56 for non-odontogenic cysts as well as a specificity of 0.59 for odontogenic cysts and 0.84 for non-odontogenic cysts (IoU: 0.30). The human control group achieved a sensitivity of 0.70 for odontogenic cysts, 0.44 for non-odontogenic cysts, and 0.56 for OPGs without cysts as well as a specificity of 0.62 for odontogenic cysts, 0.95 for non-odontogenic cysts, and 0.76 for OPGs without cysts. Taken together, our results show that a combined object detection and image segmentation approach is feasible in emulating the human clinical diagnostic reasoning process in classifying cystic lesions of the jaw.
ABSTRACT
OBJECTIVES: Detecting caries lesions is challenging for dentists, and deep learning models may help practitioners to increase accuracy and reliability. We aimed to systematically review deep learning studies on caries detection. DATA: We selected diagnostic accuracy studies that used deep learning models on dental imagery (including radiographs, photographs, optical coherence tomography images, near-infrared light transillumination images). The latest version of the quality assessment tool for diagnostic accuracy studies (QUADAS-2) tool was used for risk of bias assessment. Meta-analysis was not performed due to heterogeneity in the studies methods and their performance measurements. SOURCES: Databases (Medline via PubMed, Google Scholar, Scopus, Embase) and a repository (ArXiv) were screened for publications published after 2010, without any limitation on language. STUDY SELECTION: From 252 potentially eligible references, 48 studies were assessed full-text and 42 included, using classification (n = 26), object detection (n = 6), or segmentation models (n = 10). A wide range of performance metrics was used; image, object or pixel accuracy ranged between 68%-99%. The minority of studies (n = 11) showed a low risk of biases in all domains, and 13 studies (31.0%) low risk for concerns regarding applicability. The accuracy of caries classification models varied, i.e. 71% to 96% on intra-oral photographs, 82% to 99.2% on peri-apical radiographs, 87.6% to 95.4% on bitewing radiographs, 68.0% to 78.0% on near-infrared transillumination images, 88.7% to 95.2% on optical coherence tomography images, and 86.1% to 96.1% on panoramic radiographs. Pooled diagnostic odds ratios varied from 2.27 to 32,767. For detection and segmentation models, heterogeneity in reporting did not allow useful pooling. CONCLUSION: An increasing number of studies investigated caries detection using deep learning, with a diverse types of architectures being employed. Reported accuracy seems promising, while study and reporting quality are currently low. CLINICAL SIGNIFICANCE: Deep learning models can be considered as an assistant for decisions regarding the presence or absence of carious lesions.
Subject(s)
Deep Learning , Dental Caries , Dental Caries/diagnostic imaging , Dental Caries Susceptibility , Humans , Reproducibility of Results , Sensitivity and SpecificityABSTRACT
INTRODUCTION: Dental undergraduates will access the Internet searching for learning materials to complement their training; however, open access content is not generally recommended by dental schools. This study aimed to evaluate how dental students are using online video content. MATERIALS AND METHODS: Students from eight Universities (Athens, Birmingham, Brescia, Cardiff, Melbourne, Paris, Sao Paulo and Valdivia) representing three continents were invited to complete a survey on their access and learning from online videos. RESULTS: International students behave similarly when studying dental content online. Of 515 respondents, 94.6% use the Internet as a learning tool. It was observed that videos are not frequently recommended during didactic lectures (9.6%). But many students (79.9%) will use YouTube for their learning which includes clinical procedures. Students will check online content before performing procedures for the first time (74.8%), to understand what was explained in class (65.9%) or read in books (59.5%), to relearn clinical techniques (64.7%) and to visualise rare procedures (49.8%). More than half of the students do not fully trust the accuracy or the reliability of online content. This does not prevent students from watching and sharing dental videos with classmates (64.4%). The content watched is not shared with teachers (23.3%) even when it contradicts what was learnt in the school (38.2%). CONCLUSION: This study concludes that students regularly integrate open access digital resources into learning portfolios but are hesitant to inform their teachers about their viewing habits. Students wish to receive critical skills on how to evaluate the material they encounter outside their traditional learning space.
Subject(s)
Curriculum , Education, Dental , Brazil , Humans , Reproducibility of Results , StudentsABSTRACT
OBJECTIVES: To study caries prevalence/severity in 12-year-old children in Latvia and potential risk indicators. METHODS: A cross-sectional oral-health national survey of 12-year-old children was conducted in 2016. A nationally representative stratified-cluster probabilistic sample of 2,138 pupils in 92 schools was selected. Children were examined by seven calibrated examiners (kappa inter-examiner, intra-examiner scores of 0.71-0.77, 0.81-0.97, respectively) at school. Enamel-non-cavitated decay (D1), enamel cavitation (D3), dentine cavitation (D5), missing (M) or filled (F) status at the tooth (T)/surface (S) levels were evaluated, and decayed, missing, and filled (DMF) index scores for severity, along with the Significant Caries Index (SiC), were calculated. An associated caries factor questionnaire was completed by participants. RESULTS: The prevalence of caries was 98.5% for D1MFT, 79.7% for D3MFT, and 71.9% for D5MFT. The means (standard deviations) for severity were 9.2 (5.3) for D1MFT, 3.3 (3.0) for D3MFT, and 2.4 (2.4) for D5MFT, and 5.6 (2.1) for the SiC. Indicators associated with a lower risk of caries (D5MFT) were irregular dental visits (prevalence odds ratio PORâ¯=â¯0.45, 95% confidence interval (CI): 0.36, 0.56) and irregular use of mouthwashes (PORâ¯=â¯0.73, 95% CI: 0.60, 0.89). CONCLUSIONS: We found a high caries prevalence and severity in 12 year-old children in Latvia. Although the WHO target for 2010 (D5MFT ≤ 3) is met, the values for caries prevalence (D5MFT > 0â¯=â¯71.9%) and severity (D5MFTâ¯=â¯2.5) in 12-year-old Latvian children are higher than the European averages (D5MFT > 0â¯=â¯52%, D5MFTâ¯=â¯1.1).
Subject(s)
Dental Caries Susceptibility , Dental Caries , Child , Cross-Sectional Studies , DMF Index , Dental Caries/epidemiology , Humans , Latvia/epidemiology , PrevalenceABSTRACT
Proper peer review and quality of published articles are often regarded as signs of reliable scientific journals. The aim of this study was to compare whether the quality of statistical reporting and data presentation differs among articles published in 'predatory dental journals' and in other dental journals. We evaluated 50 articles published in 'predatory open access (OA) journals' and 100 clinical trials published in legitimate dental journals between 2019 and 2020. The quality of statistical reporting and data presentation of each paper was assessed on a scale from 0 (poor) to 10 (high). The mean (SD) quality score of the statistical reporting and data presentation was 2.5 (1.4) for the predatory OA journals, 4.8 (1.8) for the legitimate OA journals, and 5.6 (1.8) for the more visible dental journals. The mean values differed significantly (p < 0.001). The quality of statistical reporting of clinical studies published in predatory journals was found to be lower than in open access and highly cited journals. This difference in quality is a wake-up call to consume study results critically. Poor statistical reporting indicates wider general lower quality in publications where the authors and journals are less likely to be critiqued by peer review.
ABSTRACT
AIM: To estimate the global prevalence of early childhood caries using the WHO criteria. DESIGN: Systematic review of studies published from 1960 to 2019. DATA SOURCES: PubMed, Google Scholar, SciELO, and LILACS. Eligibility criteria were articles using: dmft-WHO diagnostic criteria with calibrated examiners, probability sampling, and sample sizes. STUDY SELECTION: Two reviewers searched, screened, and extracted information from the selected articles. All pooled analyses were based on random-effects models. The protocol is available on PROSPERO 2014 registration code CRD42014009578. RESULTS: From 472 reports, 214 used WHO criteria and 125 fit the inclusion criteria. Sixty-four reports of 67 countries (published 1992-2019) had adequate data to be summarised in the meta-analysis. They covered 29 countries/59018 children. Global random-effects pooled prevalence was (percentage[95% CI]) 48[43, 53]. The prevalence by continent was Africa: 30[19, 45]; Americas: 48 [42, 54]; Asia: 52[43, 61]; Europe: 43[24, 66]; and Oceania: 82[73, 89]. Differences across countries explain 21.2% of the observed variance. CONCLUSIONS: Early childhood caries is a global health problem, affecting almost half of preschool children. Results are reported from 29 of 195 countries. ECC prevalence varied widely, and there was more variance attributable to between-country differences rather than continent or change over time.