Skip to main content

Research and application of artificial intelligence in dentistry from lower-middle income countries – a scoping review


Artificial intelligence (AI) has been integrated into dentistry for improvement of current dental practice. While many studies have explored the utilization of AI in various fields, the potential of AI in dentistry, particularly in low-middle income countries (LMICs) remains understudied. This scoping review aimed to study the existing literature on the applications of artificial intelligence in dentistry in low-middle income countries. A comprehensive search strategy was applied utilizing three major databases: PubMed, Scopus, and EBSCO Dentistry & Oral Sciences Source. The search strategy included keywords related to AI, Dentistry, and LMICs. The initial search yielded a total of 1587, out of which 25 articles were included in this review. Our findings demonstrated that limited studies have been carried out in LMICs in terms of AI and dentistry. Most of the studies were related to Orthodontics. In addition gaps in literature were noted such as cost utility and patient experience were not mentioned in the included studies.

Peer Review reports

What is already known

Artificial Intelligence is being used extensively in high-income countries in different sectors including healthcare.

What this study adds

The scarcity of the current application and research of artificial intelligence in dentistry in low-middle income countries.

How this study might affect research, practice, or policy

Application and research in artificial intelligence pertinent to dentistry in low-middle income countries can greatly enhance practices and policies leading to better patient care.


Artificial Intelligence (AI) techniques such as Machine Learning (ML) and Deep Learning (DL), have the potential to execute complex diagnostic tasks, currently performed by dental specialists, that leads to improved diagnostic accuracy and increased efficiency. Adequately trained AI has the ability to detect teeth, identify pathologies and anomalies including dental caries, missing or lost teeth, periapical lesions, and maxillofacial abnormalities from dental radiographs without the involvement of any dentist or specialist in significantly less time and with more accuracy [1,2,3].

However, AI research and application in real life scenarios is a resource-intensive task and relies upon both supervised and unsupervised learning techniques that require properly curated and annotated high quality datasets. These specific types of datasets are difficult to procure and compile.

In addition, AI research in health care warrants specialized human resources including medical professionals who are familiar with AI terminologies, norms, and standards. These domains are not a part of any undergraduate or postgraduate at the moment curriculum and require additional specialized training and expertise [45]. Finally, AI researchers need to have access to expensive hardware such as graphic processing units (GPU) which are difficult to acquire in a resource capped environment.

A combination of all these factors may lead to limited advancement in health care AI research and its utility in Low and Middle-Income Countries (LMICs) [56] Ironically, it is in these same regions where the use of AI may have more value in terms of support and standardization of clinical judgment through data-orientated approaches [7]. Global dental health is complex and encompasses numerous hidden variables, including differences in disease prevalence between various races and ethnicities based on genetics. Therefore, the training datasets in dentistry generated from high-income countries would not accurately represent the population, features and disease patterns for LMICs. AI trained on datasets from high-income countries may thus introduce errors if applied in a differently placed population groups [8]. Failure to tune and align the model to a particular population could give rise to certain unintended consequences such as affecting fairness, introducing biases, and disrupting the appropriateness of that algorithm [9]. Therefore, it is quintessential that AI algorithms be trained upon context-specific environment, establishing their relevance and application [8].

Low-middle income countries face many challenges in terms of healthcare access and resources, especially dentistry. Investigating the role of AI in such countries relevant to dentistry can help identify the current applications and gaps to enhance and improve access of dental care. To optimise AI’s impact and relevance, it is essential to comprehend how it might be used for innovation in oral healthcare in settings with limited resources.

In order to identify the utility and development of AI pertaining to dentistry in LMICs, this scoping review intended to include all relevant scientific publications to analyze various study characteristics, with particular focus on the origin and quality of datasets used as well as any challenges linked to carrying out AI based research in LMICs. The level of maturity and integration of AI models mentioned were also be analyzed. In addition, considerations of cost-effectiveness or cost-utility analyses were identified. Any AI application having a patient interface or interaction was evaluated for recording patient experience in this regard.


This scoping review was conducted using a predetermined protocol following the PRISMA Extension for Scoping Reviews guidelines [10]. The protocol can be accessed through the Open Science Framework platform (

Search strategy

A 3-prong search strategy defining technique of interest (AI, DL, ML), the specialty of interest (Dentistry), and setting of interest (LMICs) was developed by the authors in collaboration with a medical information specialist (Librarian, Aga Khan University Hospital, Pakistan). An LMIC was defined according to World Bank Group Classification of Economies. The studies included in this scoping review were based on the criteria of low-middle income countries as set by World Bank [11].

(World Bank Country and Lending Groups – World Bank Data Help Desk.

The authors conducted a pilot search based on various combinations of key search terms to formulate the final search strategy which was subsequently used in this review.

Literature search

A comprehensive literature search was conducted from Jan 2010 to Feb 2023 to identify relevant publications in three major health sciences databases i.e., PubMed, Scopus, and EBSCO Dentistry & Oral Sciences Source. Furthermore, a manual search was done by the authors in Google Scholar and IEEE Xplore databases to identify relevant literature not present in the aforementioned databases.

Search terms

The search terms used for this purpose were; “Dentistry” OR “Dental AND Artificial Intelligence” OR “Deep Learning” OR “Machine Learning” AND “LMIC” OR “Low Resource setting” OR “Low and Middle Income” OR “Underdeveloped Nations” OR “Developing Countries” OR “Developing nations” OR “Economically Developing Countries” OR “Economically Developing Nations” OR “Emergent Nations”.

All primary quantitative and/or qualitative research pertaining to the implementation of AI in the context of dental health conducted in any country defined as LMIC, published in a peer reviewed journal, in English language, and published after 2010 were included. Any published protocols, conference proceedings, letters to editor, and policy documents were excluded.

Screening process

Selected citations were exported to Endnote version 20.0 (Clarivate Analytics) and duplicates were deleted. The title of each article was screened in accordance to the preset inclusion/exclusion criteria by (F.U, S. A and A.L). Any disagreement in this regard was resolved by discussion between the authors. The data from the selected studies was extracted on a predetermined proforma to chart characteristics of the included studies and key findings. These included country of origin, field of dentistry where AI was applied, dataset types and sources, annotation of dataset, description of annotators, algorithms used, computational resource, cost utility analysis, and challenges of conducting AI centered research in LMICs, if mentioned.

Eligibility criteria

The following inclusion criteria was applied:

  1. 1)

    Primary studies

  2. 2)

    Studies related to dentistry

  3. 3)

    Studies utilizing artificial intelligence models such as machine learning and deep learning.

  4. 4)

    Studies published till date

The following exclusion criteria was applied:

  1. 1)

    Reviews, editorials, commentaries, and conference papers

  2. 2)

    Studies published in languages other than English

  3. 3)

    Non indexed studies

Data synthesis

In this scoping review, we have attempted to provide a comprehensive overview of the included studies. This involved examining various relevant aspects such as the origin of research, source of datasets, study designs, datasets curation and annotation, performance metrics, and the maturity of AI algorithms used. However, it should be noted that these studies varied significantly in terms of their focus of research question/task, type of input data, model architecture, and other factors of interest. As a result, we conducted a narrative synthesis to effectively summarize the findings and provide insight into the current state pertinent to AI in LMICs.


A total of 1578 articles were initially identified following a detailed search using electronic and manual literature search platforms. After the removal of duplicates, 1357 articles underwent further screening. 27 articles were excluded on the basis of irrelevancy. A total of 1330 studies underwent a screening process for titles and abstracts. Conference proceedings, commentaries, editorials, irrelevant titles, studies not from LMICs, description of products, and reviews were excluded. Additionally, one study obtained from gray literature through Google Scholar was identified by using relevant search terms. After thoroughly scrutinizing the studies, 25 articles fulfilled the eligibility criteria. Hence, these articles were subjected to final analysis. The screening process used for this study is presented in the PRISMA flow chart (Fig. 1).

Fig. 1
figure 1

PRISMA flow chart

The overall characteristics of the included studies as extracted on the specially designed data extraction form are given in Table 1.

Table 1 Summary characteristics of the studies

Most of the studies were from India (n = 12, 48%), followed by Pakistan (n = 3, 12%) and Iran (n = 3, 12%), as illustrated in Fig. 2.

Fig. 2
figure 2

Low-middle income countries where AI has been utilized

The majority of studies from India were focused on prediction (n = 14, 56%), particularly for gender, as well as identification of pathologies such as dental caries and periapical lesions.

Overall, Orthodontics was the specialty where AI was mostly applied (n = 7, 28%), followed by Endodontics (n = 6, 24%). The majority of studies were based on validation (n = 22, 88%), and those utilizing a quantitative approach (n = 20, 80%). The highest number of studies focused on diagnostic test accuracy (n = 14, 56%), followed by prediction (n = 5, 20%) and identification of various entities (n = 3, 12%). The most frequently used type of dataset was Orthopantomograms (OPGs), used singularly or in combination with other types of data modalities. Most studies used local datasets for AI algorithm generation or for model training (n = 19, 76%).

Based on the stages of clinical AI development maturity, it was seen that most of the studies (n = 19, 76%) reached stage 2, where the authors had trained and tested AI models on actual clinical datasets. Only one study was at stage 4, where the developed model for achieving differential diagnosis was tested in the actual clinical setting.

Regarding limitations faced by the authors of the included studies, none were mentioned particularly in the context of difficulties or hurdles in carrying out the research in a LMIC setup. The limitations were generic, being confined to the size of the dataset and hence with subsequent limitations related to generalizability of the results. There was also no mention of cost-utility analysis in any of the included studies.

Heat map

The heat map generated for the type of data sets used in the included studies (Fig. 3) against the dental specialties where AI was applied indicated that the highest number of data sets used in Restorative Dentistry were those that were classified as ‘others’. These included health records, Cystatin S levels of saliva, histology patches as well as scientific papers. Lateral Cephalograms and intra-oral photographs were only used as datasets in the specialty of Orthodontics. Orthopantomograms (OPGs) were the most frequently used datasets utilized across specialties including Orthodontics, Operative Dentistry, Oral and Maxillofacial Surgery, Periodontology, and Oral radiology, with the highest number of this type of dataset used in Oral Radiology. For Endodontics, the datasets used were periapical radiographs, followed by CBCT and case difficulty assessment forms. The specialty of Orthodontics overall used a greater variety of datasets including intra-oral photographs, lateral cephalograms, OPGs, dental clinic data, and casts. This indicates that the specialists in this field are trying to develop AI applications based on a multitude of diagnostic and assessment tools.

Fig. 3
figure 3

Heatmap of the included studies depicting datasets, specialties, and sample size


AI is increasingly being adopted in dentistry in recent times, with particular focus on research in this domain [36]. However, even though AI can be considered a global phenomenon, there has been a dearth in identifying the prospects and contribution of researchers from developing countries in this field of innovation. This omission could potentially lead to biases and health inequities if AI perspectives of underdeveloped nations are not considered, where application potential may be far greater than in developed nations [37]. AI based entities could help provide patients a diagnosis, predict the spread of disease and outbreak predictions when applied in the healthcare sector within LMICs [4]. However, since there was limited reporting of its application in dentistry, therefore, this review was designed in order to encompass the focus and progress of research pertaining to dentistry and AI contributed by low and middle income countries (LMICs). A comprehensive compilation of data from these studies could be generated, highlighting dental specialties which are evolving through the use of AI, domains of research, types, and maturity of AI applications as well as recognize gaps in research. Any challenges put forth by researchers relevant to the development and application of AI would be analyzed, providing evidence for the support and progression of AI based dentistry in LMICs through relevant global agencies. This will enable researchers to focus on these gaps and improve the inclusivity and diversity of AI research in dentistry.

The LMICs category consists of 54 countries, as reported by the World Bank ( Despite our best efforts, only 25 studies from eight LMICs that met our inclusion criteria could be found. This number is significantly lower than the 168 studies found in the most recent comprehensive scoping review, which conducted its search in May 2021 [36]. It is important to acknowledge that the low number of included studies our review is consistent with the broader trend in the healthcare industry, where only 1–3% of research involves investigators from LMICs [38].

Most of the included studies were from India. This observation is consistent with the pattern of AI research in other healthcare domains, where Indian researchers have made a significant contribution [39].

The heat map indicated that Orthodontics and Endodontics were the most common specialties where AI was being applied in LMICs, with dental caries and periapical lesions being the most frequently studied pathologies. The studies mainly used validation methodology with a quantitative approach, and focused diagnostic test accuracy, which is an important domain for LMICs, since accurate and timely diagnosis could result is significant decrease in morbidity, mortality, and expenditure in a resource deprived area. The most commonly used type of data set was orthopantomograms (OPGs), used alone or in combination with other types of data sets. This finding is also in accordance with other scoping reviews [40].

The prevalence of deep learning techniques, specifically Convolutional Neural Networks (CNNs), was found to be prominent in AI algorithms, with a diverse range of CNN architectures being utilized. This dominance could be because CNNs are highly effective in addressing computer vision tasks. Classical machine learning approaches were also observed to be widely adopted, primarily in developing decision support systems, and are consistent with other literature [341]. Furthermore, datasets used in the included studies were relatively small and not representative of the broader population. The small dataset sizes used could be because most studies did not report sample size calculations. Without proper sample size estimations, researchers may not have had a clear understanding of the required number of datasets or diagnostic documentation needed to adequately represent the target population [42]. This lack of planning may have resulted in smaller datasets being used, which can impact the statistical power and generalizability of the results [1642]. This challenge is not unique to dental artificial intelligence (AI) research in LMICs, rather it has been widely documented in dental studies even from developed and resource rich countries [5364344]. Poor quality and deficient datasets can lead to problems such as biases in AI algorithms and may impact the accuracy and reliability of AI in dentistry. The lack of quantity and quality of datasets from LMICs impedes their generalizability, even for these countries, let alone for use in high-income countries or regions with different demographics.

In addition, the curation and annotation methods of datasets used in the included studies were not consistently reported. To prevent biases and inaccuracies in AI algorithms, it is essential to train them using context-specific environments. The above narrative suggests that the datasets used in AI research in dentistry are relatively small, not representative of the broader population, and the reporting of curation and annotation methods of these datasets was deficient. These issues are particularly prevalent in LMICs. This lack of transparency raises questions about the quality and reliability of the datasets, which are critical for development and validation of AI algorithms in dentistry [33645]. Therefore, it is important for researchers in LMICs to focus on sourcing high-quality, representative datasets, curation, and annotation methods to improve the development and validation of AI algorithms in dentistry, even though it be challenging and time-consuming in a resource-limited setting.

Without simultaneously addressing the fundamental issue of inadequate infrastructure, the potential of these technologies in global health remains uncertain [541]. It seems ironic that the lack of adequate resources prevent the development and implementation of AI applications in LMICs, where through the use of AI, healthcare could eventually become more cost-effective. This fact should be realized by funding and research agencies globally, so sufficient funds and resources could be directed towards for researchers in LMICs researchers to explore the various avenues of AI application in healthcare in these countries.

Although there is a burgeoning interest in the use of AI in dentistry, its translation from academic investigation to practical software application remains a work in progress. To better understand the deployment of AI in LMIC, we also stratified the included studies in our review according to the maturity of their development stages, as described by Zhang and colleagues [38]. This model describes four stages: math into algorithm, algorithm into model, model into device, and device into practice. Majority of research in AI in dentistry was focused on the algorithm-to-model stage, involving testing performance of models on labeled datasets. Only three studies progressed to the model-into-device stage. One such study, conducted by Mariam et al., used gaze-based annotation of histology images to compare the efficiency of AI versus human annotation finding AI to be more efficient [16].

The other two studies compared WebCeph, an online orthodontic software tool that uses AI technology for cephalometric analysis. Both studies utilized indigenous datasets and compared manual tracing with WebCeph tracing, but did not find any statistically significant differences in the tracing accuracy [32]. However, WebCeph was significantly faster. Notably, the authors of this review were unable to obtain details about the algorithm and dataset used to train the WebCeph model. Finally, one study utilized a machine learning algorithm to improve patient recall rate by 15% [29]. Different studies have been conducted to enhance the field of orthodontics, such as a study by Bahrami et al., emphasizing on Smart Orthodontic Brackets where utilization of machine learning algorithms can help improve and optimize orthodontic treatment outcomes [46].

Only two studies have attempted to validate the deployment of the device in a real-world environment. One study by Ehtesham et al. used AI as a decision support system for oral pathology detection. It is worth reporting that the deployment period lasted only 5 days, and the validation was conducted on a small sample of 39 patients [14].

Artificial intelligence (AI) holds promise in enhancing workflow and efficiency within healthcare practices. As such, it is essential that the AI system is user-friendly and compatible with existing infrastructure for effective implementation. However, this study had limitations, including a short duration of observation and limited number of patients, and was not entirely representative of the true environment for AI implementation and therefore warrants further research.

The other study by Ghosh et al. used bespoke messages, based on patients treatment needs, to send reminders for follow up visits which helped improve patient recall rate. This was the only study that described the patient experience with AI implementation in dental clinics [29].

In terms of future-orientated approach, there is an essential need to develop regulations concerning patient safety, confidentiality and privacy when AI is actually implemented in the real world settings. Moreover, there was a lack of cost-effectiveness or cost-utility analyses among the studies, highlighting a crucial gap in research that needs to be evaluated further. Besides the use of AI in clinical dentistry, its utility for academics and training of dental students is also an avenue worth exploring.

This study is unique since it attempts to holistically identify the research that has been conducted and subsequently published pertinent to the application of AI in dentistry hailing from LMICs. The authors of the included studies included did not mention any specific limitation in conducting AI based research in LMICs in terms of constraint of resources. However, the fact that only 25 studies from around the globe could be included in this review is a testament to the fact that the ongoing research in LMICs is significantly deficient. This could be attributed to the dearth of resources, among other possible factors which need to be further investigated. We found all included studies to have mentioned the deep learning techniques, algorithms, and Al model as well as datasets that they used in their research. However, another study done on the AI application in the healthcare sector in LMICs reported half of the included studies did not mention this information, even though the review focused on research done on real world application of AI [47].

Some limitations that were faced in conduction of this review included the difficulty in making specific and clear-cut conclusions from results of the included studies because of variability in study methodologies, reporting of the results, and the AI entities used. Therefore, a summary chart was included in order to display the results of the included studies.


This scoping review identified all pertinent literature related to AI application and research in dentistry, hailing from LMICs. It was found that most studies were from India, done in the specialty of Orthodontics, focusing on diagnostic test accuracy with validation study design and a quantitative approach. Most frequent datasets used were OPGs, with limited number of datasets in all included studies. Majority of the studies reported AI maturity at level 2. There was no reporting of limitations that researchers faced based on the lack of resources in LMICs but the scant research available from LMICs in this domain is a clear indication of this. There was no mention of cost-utility analysis or patient experience, which are gaps that need to be addressed in subsequent studies. In addition, there was a great variability in reporting of results. Overall, it was seen from the limited number of studies included in this review that the utility and research relevant to AI in dentistry is significantly deficient in LMICs. With the interest and support of the global community, researchers in LMICs would be able to attain the necessary resources to expand the application and utility of AI in dentistry in these countries, resulting in a greater number of research publications.

Availability of data and materials

All data generated or analyzed during this study are included in this published article [and its supplementary information files].


  1. Adnan N, Khalid WB, Umer F. An artificial intelligence model for teeth segmentation and numbering on Orthopantomograms. Int J Comput Dent. 2023;

  2. Umer F. Could AI offer practical solutions for dentistry in the future? BDJ Team. 2022;9:26–8.

    Article  Google Scholar 

  3. Umer F, Habib S. Critical analysis of artificial intelligence in endodontics: a scoping review. J Endod. 2022;48:152–60.

    Article  PubMed  Google Scholar 

  4. Brian W, Aline C-G, Stefan G, Nina RS. Artificial intelligence (AI) and global health: how can AI contribute to health in resource-poor settings? BMJ Glob Health. 2018;3:e000798.

    Article  Google Scholar 

  5. Mohammad-Rahimi H, Rokhshad R, Bencharit S, Krois J, Schwendicke F. Deep learning: a primer for dentists and dental researchers. J Dent. 2023;130:104430.

    Article  PubMed  Google Scholar 

  6. Jieun L, et al. Interventions to improve district-level routine health data in low-income and middle-income countries: a systematic review. BMJ Glob Health. 2021;6:e004223.

    Article  Google Scholar 

  7. Naseem M, Akhund R, Arshad H, Ibrahim MT. Exploring the potential of artificial intelligence and machine learning to combat COVID-19 and existing opportunities for LMIC: a scoping review. J Prim Care Commun Health. 2020;11:2150132720963634.

    Article  Google Scholar 

  8. Istasy P, et al. The impact of artificial intelligence on health equity in oncology: scoping review. J Med Internet Res. 2022;24:e39748.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Pethani F. Promises and perils of artificial intelligence in dentistry. Aust Dent J. 2021;66:124–35.

    Article  CAS  PubMed  Google Scholar 

  10. Tricco AC, et al. PRISMA extension for scoping reviews (PRISMA-ScR): checklist and explanation. Ann Intern Med. 2018;169:467–73.

    Article  PubMed  Google Scholar 

  11. World Bank. Low & middle income [Internet]. 2024 [cited 2024 Feb 7]. Available from:

  12. Talaat S, et al. The validity of an artificial intelligence application for assessment of orthodontic treatment need from clinical images. Semin Orthod. 2021;27:164–71.

    Article  Google Scholar 

  13. Khazaei M, Mollabashi V, Khotanlou H, Farhadian M. Sex determination from lateral cephalometric radiographs using an automated deep learning convolutional neural network. Imaging Sci Dent. 2022;52:239–44.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Ehtesham H, et al. Developing a new intelligent system for the diagnosis of oral medicine with case-based reasoning approach. Oral Dis. 2019;25:1555–63.

    Article  PubMed  Google Scholar 

  15. Koopaie M, Salamati M, Montazeri R, Davoudi M, Kolahdooz S. Salivary cystatin S levels in children with early childhood caries in comparison with caries-free children; statistical analysis and machine learning. BMC Oral Health. 2021;21:650.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Mariam K, et al. On smart gaze based annotation of histopathology images for training of deep convolutional neural networks. IEEE J Biomed Health Informat. 2022;26:3025–36.

    Article  Google Scholar 

  17. Fatima A, et al. Deep learning-based multiclass instance segmentation for dental lesion detection. Healthcare (Basel). 2023;11

  18. Bharathi M, Prasad D, Venkatakrishnamoorthy T, Dharani M. Diabetes diagnostic method based on tongue image classification using machine learning algorithms. J Pharmaceut Negat Results. 2022;13:1247–50.

    Google Scholar 

  19. Patil V, et al. Artificial neural network for gender determination using mandibular morphometric parameters: a comparative retrospective study. Cogent Eng. 2020;7:1723783.

    Article  Google Scholar 

  20. Prasad J, et al. Machine learning predictive model as clinical decision support system in orthodontic treatment planning. Dent J (Basel). 2022;11

  21. Katyal D, Balakrishnan N. Evaluation of the accuracy and reliability of WebCeph – An artificial intelligence-based online software. APOS s in Orthodont. 12

  22. Sherly Alphonse A, Kumari SV, Priyanga PT. Caries detection from dental images using novel maximum directional pattern (MDP) and deep learning. IJEER. 2022;10:100–4.

    Article  Google Scholar 

  23. Benakatti VB, Nayakar RP, Anandhalli M. Machine learning for identification of dental implant systems based on shape - a descriptive study. J Indian Prosthodont Soc. 2021;21:405–11.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Khan M, Jindal M. Multilayer perceptron to assess the impact of anatomical risk factors on traumatic dental injuries: an advanced statistical approach of artificial intelligence in dental traumatology. J Orofac Sci. 2022;14:28–34.

    Article  Google Scholar 

  25. Yadalam PK, et al. Machine learning predicts patient tangible outcomes after dental implant surgery. IEEE Access. 2022;10:131481–8.

    Article  Google Scholar 

  26. Sherwood AA, et al. A deep learning approach to segment and classify C-Shaped Canal morphologies in mandibular second molars using cone-beam computed tomography. J Endod. 2021;47:1907–16.

    Article  PubMed  Google Scholar 

  27. Mallishery S, Chhatpar P, Banga KS, Shah T, Gupta P. The precision of case difficulty and referral decisions: an innovative automated approach. Clin Oral Investig. 2020;24:1909–15.

    Article  PubMed  Google Scholar 

  28. Moidu NP, Sharma S, Chawla A, Kumar V, Logani A. Deep learning for categorization of endodontic lesion based on radiographic periapical index scoring system. Clin Oral Investig. 2022;26:651–8.

    Article  PubMed  Google Scholar 

  29. Ghosh A, Saha AP, Saha S, Das A. Promoting the importance of recall visits among dental patients in India using a semi-autonomous AI system. Stud Health Technol Inform. 2022;293:85–92.

    Article  PubMed  Google Scholar 

  30. Fidya F, Priyambadha B. Automation of gender determination in human canines using artificial intelligence. Dent J. 2017;50:116–20.

    Article  Google Scholar 

  31. Widyaningrum R, Candradewi I, Aji N, Aulianisa R. Comparison of multi-label U-net and mask R-CNN for panoramic radiograph segmentation to detect periodontitis. Imaging Sci Dent. 2022;52:383–91.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Mahto RK, Kafle D, Giri A, Luintel S, Karki A. Evaluation of fully automated cephalometric measurements obtained from web-based artificial intelligence driven platform. BMC Oral Health. 2022;22:132.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Thanh MTG, et al. Deep learning application in dental caries detection using intraoral photos taken by smartphones. Appl Sci. 2022;12:5504.

    Article  CAS  Google Scholar 

  34. Tuan TM, Ngan TT, Son LH. A novel semi-supervised fuzzy clustering method based on interactive fuzzy satisficing for dental x-ray image segmentation. Appl Intell. 2016;45:402–28.

    Article  Google Scholar 

  35. Ngoc VTN, Viet DH, Anh LK, et al. Periapical lesion diagnosis support system based on X-ray images using machine learning technique. World J Dent. 2021;12:189–93.

    Article  Google Scholar 

  36. Arsiwala-Scheppach LT, Chaurasia A, Müller A, Krois J, Schwendicke F. Machine learning in dentistry: a scoping review. J Clin Med. 2023;12:937.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Washington V, Franklin JB, Huang ES, Mega JL, Abernethy AP. Diversity, equity, and inclusion in clinical research: a path toward precision health for everyone. Clin Pharmacol Ther. 2023;113:575–84.

    Article  PubMed  Google Scholar 

  38. Zhang J, et al. An interactive dashboard to track themes, development maturity, and global equity in clinical artificial intelligence research. Lancet Digit Health. 2022;4:e212–3.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Guo Y, Hao Z, Zhao S, Gong J, Yang F. Artificial intelligence in health care: bibliometric analysis. J Med Internet Res. 2020;22:e18228.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Schwendicke F, Golla T, Dreher M, Krois J. Convolutional neural networks for dental image diagnostics: a scoping review. J Dent. 2019;91:103226.

    Article  PubMed  Google Scholar 

  41. Shan T, Tay FR, Gu L. Application of artificial intelligence in dentistry. J Dent Res. 2021;100:232–44.

    Article  CAS  PubMed  Google Scholar 

  42. Willemink MJ, et al. Preparing medical imaging data for machine learning. Radiol. 2020;295:4–15.

    Article  Google Scholar 

  43. Berisha V, et al. Digital medicine and the curse of dimensionality. NPJ Digit Med. 2021;4:153.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Leite AF, Vasconcelos KF, Willems H, Jacobs R. Radiomics and machine learning in Oral healthcare. Proteomics Clin Appl. 2020;14:e1900040.

    Article  CAS  PubMed  Google Scholar 

  45. Ma J, et al. Towards trustworthy AI in dentistry. J Dent Res. 2022;101:1263–8.

    Article  CAS  PubMed  Google Scholar 

  46. Bahrami R, Bahrami R. A new model for managing tooth movement: smart orthodontic brackets in the internet of dental things (IoDT). Med Hypotheses. 2023;179:111168.

    Article  Google Scholar 

  47. Ciecierski-Holmes T, Singh R, Axt M, Brenner S, Barteit S. Artificial intelligence for strengthening healthcare systems in low- and middle-income countries: a systematic scoping review. Npj Digit Med. 2022;5:162.

    Article  PubMed  PubMed Central  Google Scholar 

Download references




This study did not receive any funding.

Author information

Authors and Affiliations



F.U: Conceptualization, Study Design, Data Collection, Data Analysis, Writing - Original Draft, and Final Approval of the manuscript. S.A: Conceptualization, Study Design, Data Collection, Data Analysis, Writing - Original Draft, and Critical Review of the manuscript. A.L: Data Collection, Data Analysis, Writing - Original Draft, and Initial Review of the manuscript.

Corresponding author

Correspondence to Abhishek Lal.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Umer, F., Adnan, S. & Lal, A. Research and application of artificial intelligence in dentistry from lower-middle income countries – a scoping review. BMC Oral Health 24, 220 (2024).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: