C.E. Credit. Innovating Dental Education with Artificial Intelligence

ABSTRACT

Innovation is at the heart of dental education, dental practice, and optimal oral health care delivery. This paper explores the opportunities and challenges of incorporating artificial intelligence (AI) in dental education. Although the scientific reporting of AI-related research in dental education is limited, the application of AI in academia and dentistry has far-reaching implications in complementing human tasks such as radiological examinations, clinical diagnoses, anatomy training, orthodontics, and prosthodontics education. Moreover, AI has the potential to assess students’ comprehension based on automated facial recognition. Sooner or later AI will change most methods of communication, which will necessitate dental schools to prepare the next generation of students to learn about these powerful tools. Its adoption has great potential for enhancing educational experiences and healthcare as well as paving the way for preparing future dentists to apply AI techniques in clinical practice. The dental community needs to implement recommended guidelines to effectively digitalize dental education with AI technologies.

Continuing Education Credit Available: The practice worksheet is available online in the supplemental material tab for this article. A CDA Continuing Education quiz is online for this article: https://www.cdapresents360.com/learn/catalog/view/20.

KEYWORDS:

• Artificial intelligence
• dental education
• teaching
• technology

Introduction

First coined in the 1950s, artificial intelligence (AI) is “the science and engineering of making intelligent machines.¹ ” AI refers to programs and machines that can create defined rules, learn from experience, and accomplish tasks often requiring human intelligence.² A new generation of algorithms that use large datasets to evaluate and predict has fostered a recent interest in AI, namely machine learning (ML), deep learning (DL), artificial neural networks (ANN), and convolutional neural networks (CNN).³ ML is a branch of AI that uses algorithms to learn and improve from training data and predict outcomes.⁴ Digital Twins also permeate the dental field, using ML algorithms to process large quantities of data and make predictions (Table 1).⁵ DL is a subfield of ML that consists of highly interconnected structures resembling human brain cells called artificial neural networks (ANN).⁶ It possesses many simple processing elements called neurons arranged in different layers in the web, namely an input layer, an output layer, and one or more hidden layers.⁷ CNN is a subset of DL that is used especially for image processing and analysis of radiological datasets.⁸ Many people are not very familiar with the concept of AI, but despite its widespread lack of familiarity, AI is revolutionizing every walk of education and healthcare. The number of studies involving AI is growing at an exponential rate, but its potential clinical uses are still being explored.⁹

Table 1. Comparison between different AI models commonly used in dentistry and dental education. The ✓ demonstrates an identified capability for given criteria and the X demonstrates that the given technology lacks that feature. Abbreviations: ML, machine learning; DL, deep learning; ANN, artificial neural network; CNN, convolutional neural network.

Criteria for comparison	Machine Learning	Deep Learning	Artificial Neural Network	Convolutional Neural Network
Operational Definitions	Uses algorithms to learn and improve from training data and predict outcomes	Subfield of ML built on a deep hierarchy of layers	Consists of highly interconnected structures resembling human brain cells	Subfield of DL used for processing visual and other two-dimensional data
Autonomous learning	X	✓	✓	✓
Composed of layers	X	✓	✓	✓
Volume of data	Small and medium-sized datasets	Large datasets	Large datasets	Large datasets
Human intervention	✓	X	X	X
Role in Dentistry	Commonly used in many aspects of dental education such as oral diagnosis and dental imaging	Commonly used for identifying dental caries and image processing	Commonly used in orthodontic treatment planning and endodontics	Promising ability in endodontics, oral pathology, and diagnostic imaging

The use of AI is evident in handling large volumes of data that would otherwise take a human too long or simply be impossible to process.¹⁰ In healthcare, AI is being used in patient monitoring, treatment recommendations, imaging and radiology, and medical record keeping.¹¹ AI has innovated the way we practice medicine, and the COVID-19 pandemic has only accelerated this transformation in dentistry. AI is already being applied to a variety of dental activities, including orthodontics,^12,13 dental imaging and radiology,^14,15 periodontology,^16,17 and cariology.^6,18 In endodontics, AI technology is being developed to detect periapical lesions and vertical root fractures via a DL and CNN learning model, respectively.^19–21 AI technology is also being used to automate the process of denture design, as the existing computerized design process requires tedious manual inputs and multiple denture fitting procedures.²² By automating the handling of information and data, dentists will be able to focus on other activities.

While AI frameworks in dentistry are not something unheard of, the application of AI to dental education systems remains largely unexplored. Current dental education does not prepare future dentists to incorporate AI technologies into their practice.²³ Curriculums for dental education need to extend beyond courses that teach basic and clinical sciences by including material to improve dental students’ ability to practice in an AI-driven environment.²⁴ Increasingly, routine cognitive tasks can be performed by machines, which will require dental schools to refine and improve their education curricula to incorporate AI into their day-to-day operations.² This paper aims to provide various applications of AI in dental education by answering the following questions: 1. what are the different AI techniques used in education? 2. How is AI currently being utilized in dental education (basic sciences, preclinical sciences, and clinical sciences)? 3. What challenges and limitations are considered with AI pedagogical tools?

Materials and Methods

Inclusion and Exclusion Criteria

The review included both gray literature and peer-reviewed journal articles that were published in English between January 2009 and March 2023. Due to the vast amount of research available on the application of AI in dentistry, the review focused specifically on dental education. This included the applications of AI in teaching theoretical, preclinical, and clinical dental knowledge at the undergraduate and graduate levels. Incomplete studies, letters to editors, conference readings, and posters were excluded from the review.

Search Strategy

The search strategy aimed to identify all relevant studies pertaining to AI in dental education. To achieve this, a thorough search was conducted across four databases (PubMed, Scopus, Web of Science, and Cochrane Library) and gray literature. Databases were searched on March 20, 2023 with the following search terms: “artificial intelligence” AND “dental education,” “artificial intelligence” AND “dental students,” “artificial intelligence” AND “dental educators.” Two reviewers independently assessed article eligibility based on the established inclusion and exclusion criteria, with a third reviewer available to resolve any discrepancies that arose. The references were then transferred to EndNote, where duplicate citations were eliminated.

AI in Education

Machine Learning (ML)

ML is an established concept, but its advancements in dental education curricula contain very limited information regarding its practices. ML is a branch of AI with algorithms that enable systems to learn autonomously and improve from experience without human assistance.²⁵ In this sense, virtual patients with an AI chatbot were created to help dental students develop their diagnostic skills.²⁶ Additional AI pedagogical tools are summarized in Table 1.

Deep Learning (DL)

DL is a form of ML that consists of artificial neural networks built on a deep hierarchy of layers, with each layer solving different pieces of a complex problem.²⁷ As a pedagogical tool, DL has been applied in the identification of dental caries and oral lesions from digital radiographs and other digital images.²⁸

Neural Networks

ANN is an AI modeling technique consisting of a highly interconnected structure resembling human brain cells.⁷ Previous studies show that an ANN can be trained to interpret dental data accurately without requiring prior knowledge. Some examples of its application include: locating the minor apical foramen,²⁹ determining the working length,³⁰ and determining the effects of trace elements on the mechanical properties of dentin.³¹ CNNs are referred to as deep learning structures and the diagnostic accuracy of these models is approaching levels of human expertise.³² The integration of these models in the day-to-day operations of dental schools would bring dental education to a new era of digitalization.

Digital Twins

With the increasing ability to capture large amounts of evolving data, the concept of Digital Twins enables the dynamic convergence of the physical and digital worlds.³³ In simpler terms, Digital Twins technology allows for real-time monitoring and analysis of complex systems, providing valuable insights that can improve our understanding and decision-making.³⁴ In dentistry, it can help in analyzing and predicting complex dental issues, leading to better treatment and outcomes.⁵

AI in Dental Education

Dentistry has seen major changes over the past decade, but dental education still remains largely traditional.³⁵ The process of becoming a dentist requires extensive training, which is often centered on six domains: knowledge, technical skills, critical thinking, ethics, social responsibility, and interprofessional education/practice.³⁶ Dental students are required to become competent in these areas as defined by the Commission on Dental Accreditation (CODA) and the American Dental Education Association (ADEA).³⁷ A large part of the dental school curriculum involves absorbing a great deal of information and applying it to patient care. It was noted that dental schools face difficulty making curriculum changes that adopt new technologies, which could explain the small amount of time dedicated to familiarizing dental students with novel technologies such as AI. Survey studies clearly demonstrate that dental students currently lack education and knowledge of AI, which is essential to practice in the 21^st century.²³ However, the implementation of AI seems inevitable since increasingly more papers are being published on how AI should be adopted in dental education.^2,9 With the growing use of AI in dentistry, the American Dental Association Standards Committee on Dental Informatics recently released a white paper on the use of artificial and augmented intelligence in dentistry.³⁷ Several successful initiatives for integrating AI in dental training are presented in Table 2.

Table 2. Initiatives for AI in dental education.

Institution	Project
New York University College of Dentistry	Practicing teeth setup placement in dentures with Dental Academy software
University of the Pacific Arthur A. Dugoni School of Dentistry	Presented Bolman and Deal’s Four Frames approach as a guide to adopting AI in dental education
Arizona School of Dentistry and Oral Health (ASDOH)	Created an AI-assisted removable partial denture design software
University of Connecticut (UConn) School of Dental Medicine	Invented an AI educational tool for orthodontists
University of Florida College of Dentistry (UFCD)	Developed a Dental AI Education Program

Traditional face-to-face learning allows teachers to respond more effectively to students since they can observe their facial expressions and immediately modify their teaching methods to meet their needs. Automatic feedback plays a critical role in improving the students’ performance, suggesting that AI technologies can be used as supplemental pedagogical tools to support students’ learning and may also be used as an indicator of students’ interactivity in the classroom. With the overflow of information, it has become a challenge for dental students to gain a comprehensive understanding of the content. Facial expression analysis may be integrated in dental education systems to better assess the students’ comprehension and adjust the course content accordingly, contributing to the widespread introduction of AI tools to assist with enhancing dental pedagogy (Figure 1). Ekman and Friesen proposed that facial expression may be divided into seven different states: happiness, surprise, sadness, fear, anger, disgust, and natural.³⁸ Similarly, Maqableh et al. used facial expressions to measure students’ interaction with distance learning environments during the COVID-19 pandemic.³⁹ As an example, happiness may be matched with a high level of understanding. DL may be used to enhance the engagement of students in digital learning along with a neural network model to further analyze the students’ facial features and automatically assess their understanding. Despite the adoption of these technologies in medical education, dental education is still in its infancy and could use improvements to its learning curriculum.

Figure 1. (a) Facial landmarks that may be used for image processing. (b). Isolating the eyes from the rest of the face. (c). Database of eyes. (d). Variations of facial expressions. (e). AI workflow in assessing the students’ comprehension with facial expression analysis.

Basic Dental Sciences

During the COVID-19 pandemic, anatomy education experienced a disruption, with remote teaching becoming widely adopted among anatomy disciplines.^40,41 Students and educators alike have called for the continued integration of AI into anatomy education as a standard teaching tool, valuing these advanced technologies over the traditional approach.⁴² AI capabilities are currently integrated into anatomy online learning systems for student monitoring, deployed through three-dimensional anatomical model development and anatomy gamification.⁴² Using AI, anatomy education and assessments could be delivered more consistently and efficiently, and errors and patterns may be identified that humans might otherwise overlook. Recently, chatbots and conversational agents have been developed to provide “Googleable” answers to anatomy questions.⁴³ Automation may relieve educators from traditional teaching tasks, allowing them to focus on more complex pedagogical tasks such as curriculum design. These recent advancements serve as a guide for how dental educators can integrate AI in clinical anatomy disciplines.

Preclinical Dental Sciences

Diagnosis is the foundation for all dental treatments. Effective and efficient diagnosis is an essential skill that dental students must develop as part of their undergraduate dental education.²⁶ Diagnostic learning may be effectively reenacted through repeated practice of clinical cases followed by faculty feedback, and more recently virtual patients are able to help dental students improve their skills.^44,45 By simulating virtual patients with AI, Suárez et al. provided dental students with multiple clinical cases to practice as well as more realistic interactions with the virtual patient, ensuring their training and adaptation to new technological developments.²⁶

Radiological examination is an integral part of patient management in dentistry, as it often assists in the diagnosis of oral pathologies.⁴⁶ With AI, dental students may be able to better detect pathologies and interpret images more quickly and accurately. Although teaching strategies may vary across different countries, an undergraduate dental student primarily studies radiation physics, radiographic techniques, and digital imaging interpretation in dentomaxillofacial radiology classes, as one example.⁹ Radiographic images as an element of big data may be gradually introduced within the curriculum, along with the rationale informing AI algorithms and possible implementations in dentistry.⁴⁷ Students can train on far more radiographs when using an AI system, rather than training on manually selected and annotated radiographs. The successful implementation of AI into dentistry will require extensive training in dental education.

Clinical Dental Sciences

The clinical aspect of dental education is still in the early stages of AI development, making it challenging to define what dental students need to learn. However, orthodontics and radiology have emerged as frontrunners in AI implementation.^48,49 The University of Connecticut (UConn) School of Dental Medicine developed an AI based model for diagnosis and treatment planning of orthodontic patients requiring extraction/non-extraction treatment, which could assist dental students with all aspects of orthodontic patient care (e.g., measurements and data interpretation).⁵⁰ Additionally, the University of Florida College of Dentistry (UFCD) recently partnered with Overjet (AI-powered technology for cavity detection and bone level measurements in dental X-rays) to introduce dental students and residents to AI technology through an education program consisting of a lecture series and hands-on learning opportunities to enhance their diagnostic and case presentation skills.⁵¹

Prosthodontics is another diverse and complex dental specialty and one that could benefit from the routine adoption of AI technologies.⁵² Students at the New York University (NYU) College of Dentistry have been practicing teeth setup placement in dentures with Dentca Academy software, an AI technology for teaching artificial teeth arrangement for a removable prosthesis course.⁵³ With the help of AI, dental students can learn how to create a more accurate denture design and implement a fast and efficient workflow for the production of dentures. Similarly, the Arizona School of Dentistry and Oral Health (ASDOH) also created an AI-assisted removable partial denture design software, which uses specialized algorithms to automatically generate denture designs for any possible missing tooth configuration and clinical scenario. Additionally, the software contains a game component that provides AI-generated clinical scenarios and instant feedback to the students.⁵⁴

Challenges with AI

Although AI holds many promises, it also has its own set of challenges and limitations that cannot be overlooked. Insufficient time in curricular hours has led to challenges in adjusting new content areas. Studies have also concluded that the primary concern with AI diagnosis and treatment is anonymizing confidential patient information to ensure secure data sharing among different institutions.⁷ Moreover, dental schools currently lack the faculty expertise required to teach AI in dental content. The lack of expertise in teaching AI to dental students may be resolved by promoting interfaculty collaboration between healthcare experts, engineers, and computer scientists.⁵⁵ There is also a need to balance the teaching and learning dental students receive from human experts and machines, as AI should be incorporated as a complement to dental educators. Moreover, teaching communication, ethics, and leadership skills in dentistry are likely to grow in importance with the adoption of AI in dental education.^2,48,56,57

Consideration should also be given to assessment measures from AI.^2,57 To determine the feasibility of integrating AI technology into a course, a pilot implementation can be conducted to determine student and faculty perceptions, as well as pilot outcome assessments. A successful pilot, along with positive perceptions and learning outcomes, could pave the way for the future expansion of an AI initiative at a dental school.^2,57 The current curriculum needs input from experts in dentistry and education who will need to work closely with data scientists to ensure that AI tools are integrated correctly.

Conclusion

Thus far, only a limited number of studies have reported findings about AI in dental education. In this paper, we highlight the usefulness of incorporating AI into the day-to-day operations of dental schools and dental education curricula. The current educational trend specifically concentrates on using AI models to help dental students with radiology training, orthodontics, denture design, diagnostic skills, and anatomy education, with several successful initiatives at UConn School of Dental Medicine, UFCD, NYU College of Dentistry, ASDOH, and University of the Pacific Arthur A. Dugoni School of Dentistry. AI, which complements tasks that would otherwise require human intelligence, has great potential for enhancing educational experiences and healthcare delivery, as well as paving the way for preparing future dentists to apply AI techniques in clinical practice.

Acknowledgments

MAS is a recipient of the DenburTech, New Jersey Health Foundation, and TechAdvance Awards. This publication is dedicated to the memory of Dr. H. Afsar Lajevardi⁵⁸, a legendary pediatrician (1953–2015) who passed. We will never forget Dr. H Afsar Lajevardi’s kindness and support. The views expressed in this paper are those of the authors and do not necessarily reflect the views or policies of the affiliated organizations. The authors hereby announce that they have active cooperation in this scientific study and preparation of the present manuscript. The authors confirm that they have no financial involvement with any commercial company or organization with direct financial interest regarding the materials used in this study.

Supplementary Data

Supplemental data for this article can be accessed online at https://doi.org/10.1080/19424396.2023.2217692.

Additional information

Notes on contributors

Mohammad Ali Saghiri

MA Saghiri was a visiting scholar at the University of Michigan School of Dentistry where he published more than 20 articles under the supervision of the late Prof. Kamal Asgar. He entered academia as an Assistant Professor of Dental Materials at Azad Dental School in 2012 and later was appointed as an Associate Professor and head of IT at Azad Dental School. Dr. Saghiri returned to the US in 2013 and joined Prof. Nader Sheibani’s Lab as an Assistant Scientist at the Department of Ophthalmology and Visual Sciences. After the completion of his postdoctoral studies, he joined Philips Inc. as a Clinical and Biological Scientist and mostly worked on cutting-edge technology for oral hygiene care until 2018. Presently, he is the Director of the Biomaterials Research Center and an Assistant Professor in the Department of Restorative Dentistry at Rutgers School of Dental Medicine. He is also an Adjunct Assistant Professor in the Department of Endodontics at the University of the Pacific School of Dentistry. Dr. Saghiri has authored and co-authored over 120 articles in peer-reviewed dental journals that address scientific, research, educational, and clinical topics. He was awarded three times for his outstanding contributions to research and endodontics at Azad Dental School. Additionally, he received the DenburTech Award, which is only given to clinical scientists who have published more than 100 articles under 40 years of age.

Julia Vakhnovetsky

Julia Vakhnovetsky recently graduated from Rutgers University with a major in Biological Sciences and is currently the Clinical Coordinator at the Biomaterials Research Center at Rutgers School of Dental Medicine. She has authored and co-authored over 10 articles in peer-reviewed dental journals that address scientific, research, educational, and clinical topics as an undergraduate student. She will continue her graduate studies in dental medicine at the University of Michigan School of Dentistry.

Elham Samadi

Elham Samadi is an international dentist who recently joined the Biomaterials Research Lab at Rutgers School of Dental Medicine. Her focus is on the impact of various systemic diseases on the quality of dentin and the longevity of dental treatments, with a particular interest in diabetes mellitus. Additionally, she has taken courses in digital dentistry to enhance her skills and knowledge in this area.

Mahsa Amanabi

Mahsa Amanabi is an international dentist who recently collaborated with the Biomaterials Research Lab at Rutgers School of Dental Medicine. Her current research focuses on using artificial intelligence to detect diabetes through voice analysis, and she is also developing an innovative expandable endodontic sealer.

Steven M. Morgano

Steven M. Morgano is a fellow of the American College of Prosthodontists, a past president of the Massachusetts section, and has been a member of the College since 1979. He has taught at eight different dental schools, including Boston University and Harvard School of Dental Medicine, and at universities in Kuwait and Dubai. He is currently a professor and chairman of the Department of Restorative Dentistry at Rutgers School of Dental Medicine. Dr. Morgano received his DMD degree from Tufts University School of Dental Medicine, completed his General Practice Residency at Philadelphia General Hospital, and received his specialty certificate in prosthodontics from the Hines Veterans Affairs Medical Center in 1979. He practiced as a prosthodontist in the Veterans Affairs for 15 years before devoting himself to academics full-time. Dr. Morgano is also a Diplomate of the American Board of Prosthodontics and has authored and co-authored over 90 articles in numerous dental journals.