CHAPTER 39

Artificial intelligence and dental clinical practice

https://doi.org/10.56238/sevened2024.012-039

Gabriella Karolyne Pompeu Martins1, Gustavo Bezerra dos Santos Lira2, Jucivaldo Martins
Gonçalves3, Daysa da Silva Martins4, Thalles Arievo Mota Sales5, Lurdete Maria Rocha Gauch6,
Andréa Maia Corrêa Joaquim7, Camila Lima de Andrade8, Ana Carla Carvalho de Magalhães9,
Gabriel Mácola de Almeida10, Liliane Silva do Nascimento11 and Hércules Bezerra Dias12

1
Undergraduate student in Dentistry. Federal University of Pará. Institute of Health Sciences. Faculty of Dentistry. Belém,
PA, Brazil.
ORCID: 0000-0001-9153-5219
E-mail: [email protected]
2
Graduating in Dentistry. Federal University of Pará. Institute of Health Sciences. Faculty of Dentistry. Belém, PA, Brazil.
ORCID: 0000-0002-6769-6781
E-mail: [email protected]
3
Graduating in Dentistry. Federal University of Pará. Institute of Health Sciences. Faculty of Dentistry. Belém, PA, Brazil.
ORCID: 0009-0002-2970-2674
E-mail: [email protected]
4
Undergraduate student in Dentistry. Federal University of Pará. Institute of Health Sciences. Faculty of Dentistry. Belém,
PA, Brazil.
ORCID: 0009-0002-8933-4762
E-mail: [email protected]
5
Postgraduate student in Dentistry, FIFAMA College, Belém, PA, Brazil.
ORCID: 0000-0001-9836-9700
E-mail: [email protected]
6
Doctor in infectious and parasitic agents. Professor, School of Dentistry, Institute of Health Sciences, Federal University
of Pará. Bethlehem, PA. Brazil.
ORCID: 0000-0002-3750-8874
E-mail: [email protected]
7
Doctor in Dentistry (Oral Rehabilitation). Professor, School of Dentistry, Institute of Health Sciences, Federal University
of Pará. Bethlehem, PA. Brazil.
ORCID: 0000-0001-7547-6490
E-mail: [email protected]
8
Doctor in Clinical Dentistry (Oral Rehabilitation). Professor, School of Dentistry, Institute of Health Sciences, Federal
University of Pará. Bethlehem, PA. Brazil.
ORCID: 0000-0001-5665-9865
E-mail: [email protected]
9
MSc in Public Health. Professor, School of Dentistry, Institute of Health Sciences, Federal University of Pará. Bethlehem,
PA. Brazil.
ORCID: 0000-0003-0647-0786
E-mail: [email protected]
10
Master in Public Health. Federal University of Pará. Belém, PA, Brazil.
ORCID: 0000-0002-0618-829X
E-mail: [email protected]
11
Doctor in Public Health Nursing. Professor, School of Dentistry, Institute of Health Sciences, Federal University of Pará.
Bethlehem, PA. Brazil.
ORCID: 0000-0002-5943-6314
E-mail: [email protected]
12
Doctor in Dental Sciences (Restorative Dentistry). Professor, School of Dentistry, Institute of Health Sciences, Federal
University of Pará. Bethlehem, PA. Brazil.
ORCID: 0000-0002-5042-5782
E-mail: [email protected]

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
ABSTRACT
Artificial intelligence (AI) has gained notoriety in the dental field, since the emergence of several tools that
help dentists in the performance of their activities and clinical routine, contributing to the early diagnosis of
diseases, personalization of treatments, and data management. The objective of this chapter is to address the
impact of artificial intelligence on clinical dental practice, based on the literature of the last 5 years, available
in the PUBMED, SCIELO and Google Scholar databases. Intelligent systems that have the potential to
transform clinical practice in dentistry, support the professional in the quality of the diagnosis and reduce the
cognitive wear of the professional, contributing to the reduction of neglected information, leading him to the
best treatment decision. In this sense, the conscious and ethical use of AI in dentistry is essential for an
adequate and responsible integration of AI systems into the clinical routine of the dental surgeon. Although
there are limitations and flaws presented by AI resources that still need to be addressed, the advantages
presented by these tools stand out and show promise in the dental field. The development, implementation,
and receipt of AI in dentistry must be based on well-being, privacy, the law, and respect for autonomy and
decision-making.

Keywords: Artificial intelligence, Dentistry, Evidence-based clinical practice.

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
INTRODUCTION
In dental practice, the use of Artificial Intelligence (AI) has been growing rapidly, optimizing
processes, improving efficiency, and aiding decision-making through advanced data analytics. In
addition, AI assists in the early diagnosis of diseases, personalization of treatments, and data
management (AGRAWAL and NIKHADE, 2022).
It is evident that artificial intelligence has become a promising tool in dentistry (RASTEAU
et al., 2022; MIRAGALL et al., 2023). With the advancement of technology, the application of its
tools in various areas of health has become commonplace, and in the dental field this was no
different. The various tools that artificial intelligence presents have been of great help to dentists in
the performance of their activities and clinical routine (GHODS et al., 2023). Performing simple
tasks ranging from patient scheduling to more complex ones can currently be developed using AI
(THURZO et al., 2022; GHODS et al., 2023).
A relevant aspect is the use of AI in patient communication, training (continuing and
continuing education of professionals), and monitoring treatment progress, constituting a new
domain for AI implementations (THURZO, 2021). Data and combinations of data are essential to AI
processing algorithms. Therefore, the greater the range of data, the better the learning input to the
machines, enabling better decisions (HENDI et al., 2024).
The success of the use of intelligent systems in dentistry is due, in particular, to
Convolutional Neural Networks, which are deep learning algorithms capable of capturing an input
image, assigning importance (weights and biases that can be learned) to various aspects or objects of
the image and being able to differentiate one from the other. (Data Science Academy, 2022;
TANDON and RAJAWAT, 2020).
These intelligent systems have the potential to transform clinical practice in dentistry by
supporting the practitioner in the quality of diagnosis and reducing the number of images to be
manually examined. In this way, machine learning reduces the cognitive wear and tear of the
professional, contributing to the reduction of neglected information, leading to the best treatment
decision (ARSIWALA-SCHEPPACH et al. 2024; SONG et al., 2022).
There are some fears on the part of some professionals regarding the use of these
technologies, some of them being related to providing the high amount of data that AI systems need
to perform functions efficiently and accurately. Although health data are personal and sensitive and
need protection throughout the use process, the possibility of data leakage in the previous training of
these systems is an issue raised by health professionals (FATIMA et al., 2022; RASTEAU et al.,
2022).
This is in line with the guidance of the conscious and ethical use of AI in dentistry, which
must consider when to apply AI and how to use it appropriately and responsibly, as as many dentists

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
are accelerating towards the integration of AI systems into their daily clinical routine, legal and
ethical issues are becoming even more pertinent. (ROGANOVIĆ, RADENKOVIĆ and MILIČIĆ,
2023).
The incorporation of AI in the clinical practice of the most diverse dental specialties can help
in the early detection of diseases, optimization of the dental workflow, time management and cost
reduction) for patients, which should lead to treatments with better cost-benefit and high quality
resolution (WANG et al., 2020; TANDON and RAJAWAT, 2020; HENDI et al., 2024).
The objective of this chapter is to address the impact of artificial intelligence on clinical
dental practice, in its most diverse specialties, based on the literature of the last 5 years, available in
the PUBMED, SCIELO and Google Scholar databases.

APPLICATIONS IN DENTISTRY
Caries is one of the leading clinical conditions in the world and early diagnosis is critical,
especially considering the socioeconomic challenges that affect access to dental care (WANG et al.,
2020). To locate and determine the extent of caries, the dentist should integrate the clinical
examination with the complementary examination with radiographs. However, interproximal
radiographs have low sensitivity for enamel-restricted lesions (CARVALHO et al., 2021) and
primary caries on proximal surfaces (ARSIWALA-SCHEPPACH et al. 2024). The potential of
incorporating AI into restorative dentistry is noted, as early detection and treatment can preserve
tooth structure and prevent the cycle of invasive treatments (DEVLIN et al., 2021).
Due to the progressive and irreversible nature of caries injury, AI-driven diagnostic tools
potentially improve diagnostic accuracy, facilitate early detection, and support treatment decision-
making (AL-NAMANKANY, 2023). In this way, it is possible to diagnose and intervene in lesions
regardless of their stage and location, such as secondary caries, installed adjacent to the restorations,
compromising the useful life of the restorative treatment (CHAVES et al., 2024). The findings reveal
that AI algorithms may be a promising way to improve caries prediction, detection, and management
(AL-NAMANKANY, 2023).
Specially designed for image processing to capture spatial relationships in two-dimensional
input data (CHAVES et al., 2024), convolutional neural networks can be used as an aid in the
diagnosis of caries during radiographic analysis. Interproximal radiography, for example, which is
used for the diagnosis of caries, is a two-dimensional image and can be easily processed by neural
networks, optimizing clinical practice (ARSIWALA-SCHEPPACH et al. 2024).
Studies that evaluated the observation time of dentists until the diagnosis of caries showed
that professionals who were assisted by software arrived at the report in less time (DEVLIN et al.,
2021; ARSIWALA-SCHEPPACH et al. 2024). It was also observed that the convolutional neural

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
network was able to detect 92.5% of caries, providing the patient with appropriate intervention
(KÜHNISCH et al., 2022). Machine learning will gradually enable minimally invasive treatments
(GOMEZ et al., 2022) and in cases of major tooth destruction; technologies such as CAD/CAM are
used for restorations, such as inlays, onlays, crowns, and bridges (TANDON and RAJAWAT, 2020).
AI is expected to be fully incorporated into dentistry. Considering that the programs are based
on mathematical models influenced by human neurons, AI does not pose a risk to the professional
class, since it is the responsibility of dentists to foster research and feed software for full use
(TANDON and RAJAWAT, 2020). However, it is crucial to be aware of potential errors in the
interpretation of data by AI programs, and combining the technology with conventional
methodologies can minimize these errors (CABRAL et al., 2021).
Although it demonstrated the accuracy of AI in diagnosis, the study by Chaves et al. (2024)
pointed out the presence of false positives when diagnosing caries lesions on healthy surfaces,
wrongly attributing the diagnosis of secondary caries to residual caries, and erroneously classifying
ill-adapted restorations, in addition to false negatives for caries lesions restricted to enamel
(CHAVES et al., 2024). This event results from the low performance of the software in the presence
of impacted, crowded teeth, dental implants, and areas of greater contrast, such as overlapping,
enamel, external oblique crest, and pulp stones, in addition to the anatomical variability of molars
(PETHANI, 2020).
The inability of the operating system or even less theoretical deficit can lead the patient to
unnecessary or inefficient restorative treatments. Bringing to light the relevance of adequate training
for the reproducibility of new diagnostic systems by CD (DEVLIN et al., 2021). The incorporation
of AI and maintenance of the database, such as the increment of neural layers, for example, is
indispensable for greater diagnostic sensitivity (TANDON and RAJAWAT, 2020). Their products
assist in early disease detection, dental workflow optimization, time management, and cost reduction
(HENDI et al., 2024), for patients - because early detection of caries is more cost-effective than
treating advanced conditions (WANG et al., 2020).

APPLICATIONS IN ORAL PATHOLOGY

In oral cancer and precancerous conditions, AI has aided on a large scale in anamnesis and
diagnosis, a product of neural network processing, in the screening and comparison of test results,
such as MRI and CT scans, to assess potential for metastasis (AHMED et al., 2020; BONNY et al.,
2023; TANDON and RAJAWAT, 2020). With the supply of the image bank, neural networks allow
the processing of two-dimensional images, whether they are radiographs, MRIs, or even photographs
(CHAVES et al., 2024), providing sensitive and assertive diagnoses, in some cases, surpassing
human professionals in comparison efficiency (PETHANI, 2020).

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
 In terms of diagnostic by image comparison, neural networks can be trained to identify
different types of cysts, bone lesions, and oral cancer in diagnostic images; in addition to predicting
the dosage of radiation to the jaw for oropharyngeal cancer treatment, with high accuracy
comparable to human specialists (PETHANI F, 2020). Studies show the competence of unsupervised
learning models for diagnosing clinical conditions, including soft tissue calcifications, tissue tumors,
and anomalies, such as taurodontia (SONG, 2022).
Deep learning techniques support the early detection of dental anomalies, given that, if
ignored, they can lead to serious systemic complications (LEE et al., 2022). In histology, intelligent
systems also make it possible to improve diagnosis, by virtue of their application to identify
malignant character and specific features in histopathological sections, such as cell morphology,
nuclear atypia, and tissue architecture (GHODS et al., 2023; GONZÁLEZ, 2021; DHOPTE et al.,
2023 ). In addition to image processing, AI is able to predict the occurrence of pathologies, such as
squamous cell carcinoma, based on clinical and genetic data (PETHANI, 2020; DHOPTE et al.,
2023).
Intelligent systems have the potential to transform clinical practice in dentistry, in the sense
that machine learning reduces the cognitive wear and tear of the professional, contributing to the
reduction of neglected information, leading them to the best treatment decision (ARSIWALA-
SCHEPPACH et al., 2024; SONG, 2022). In addition to reducing human errors due to the
concealment of data in the anamnesis or medical history, such as the diagnosis of bisphosphonate-
related osteonecrosis before extraction (AHMED et al., 2020).
With the range of results greater than 80% in favor of AI accuracy in oral pathology,
educational institutions need to be prepared to offer ongoing AI training in dentistry (DHOPTE et al.,
2023; GHODS et al., 2023). Considering the complexity of oral anomalies, AI should be used as an
additional benefit to be controlled by the dentist, and not replace him (BONNY et al., 2023).

APPLICATIONS IN PERIODONTICS
In periodontics, Artificial Intelligence already enables the analysis of diagnostic data, such as
clinical parameters, radiographs and intraoral images, to assist in the diagnosis and classification of
periodontal diseases from electronic medical records and datasets inserted into AI software that learn
patterns and indicators of periodontal disease, assisting in decision-making and planning for the best
possible treatment (DHOPTE and BADGE, 2023).
In this specialty, in particular, international studies show continuous development of
intelligent tools and systems useful to complement the dentist's decision-making. Deep Learning
algorithms associated with Neural Networks have high potential for usefulness and accuracy in the
diagnosis and prediction of clinical conditions based on values of probing depths, insertion levels,

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
and bleeding points, extracted from patient records (BONNY et al., 2023; DHOPTE and BADGE,
2023).
Through the techniques of learning systems from a database of existing and defined
characteristics in the classification of periodontal diseases, it is possible to benefit periodontists and
patients in the accurate and quality diagnosis while saving time and resources, in addition to reducing
human error. Despite this, it is understood that these systems are still far from replacing the
knowledge of the health professional, and should be applied in a safe and complementary way to the
judgment of the dental surgeon (BONNY et al., 2023).
A 2024 primary study evaluated the accuracy of an AI algorithm trained to analyze signs of
periodontal disease through image records from 60 patients. In their results, the authors found a
sensitivity of 91%, specificity of 86% and an overall accuracy of 88% of the algorithm,
demonstrating the use of AI as a promising tool in the early identification of periodontal diseases
through intraoral imaging and favoring the reduction of diagnostic variability and support of more
timely interventions (ALAM et al., 2024).
In addition, it has also been published that AI has the ability to identify patterns and models
that may be difficult for humans to discern or that would require a significantly longer time, such as
the identification of periodontalally compromised premolars and molars with 90% and 95%
accuracy, respectively. This information is of great relevance to support treatment protocols, leading
to positive impacts on diagnostic accuracy and overall treatment outcomes (HENDI et al., 2024).
Despite the promising prospect of these tools, it is important to take into account that, as
much as they can detect pathologies and direct diagnoses accurately, it is the practitioner's treatment
recommendations based on the patient's overall context, available treatment options, and the patient's
adherence to treatment that determine the true impact on clinical outcomes (PETHANI, 2021).
Chawla et al. (2023) published a study conducted in 2023 that evaluated the understanding
and acceptability of periodontists regarding the use and application of AI in this area of expertise.
The study reveals that there was great acceptability among professionals for the use of AI in the
diagnosis of periodontal diseases, but resistance to its use in the predictability of clinical loss of
insertion.
Despite understanding the basic factors of the concept of artificial intelligence and expressing
a desire to incorporate such systems into the processes of diagnosis and therapeutic planning of
periodontal conditions, part of the interviewees believe that their use for diagnosis is not more
accurate than that of the professionals themselves to the point of replacing them (CHAWLA et al.,
2023).

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
 Also from this perspective, it is a fact that the investigation of periodontal diseases and
conditions is linked to complex contexts that require a multidimensional understanding of the factors
that are associated with their manifestation and progression.
Currently, despite the possibility of teaching an AI about the scenarios that comprise
periodontal diseases and subsidizing a positive impact on the diagnostic process, these tools are still
incapable of reproducing the nature of the doctor-patient relationship, built from bonding and
humanization. And second, in the process of integrating medical history associated with physical
examination in situations characterized by ambiguity and uncertainty, which is beyond the current
reach of AI technology (CHAWLA et al., 2023; SACHDEVA et al., 2021).
Although the development of AI within health scenarios and, more specifically, in
periodontics is still moving at a short pace, it is observed that dentists understand and visualize the
use of intelligent systems as facilitating resources for clinical practice and decision-making, assisting
professionals in the correct interpretation of oral anomalies and minimizing human error. therefore, a
modality of inevitable development and integration in dentistry (AHMED et al., 2021).

APPLICATIONS IN DENTAL IMAGING

Radiological examination is a fundamental component in the management of patients in
dentistry. It is widely used to complement and assist in the clinical diagnosis of pathologies affecting
the teeth and adjacent structures. Accuracy and the ability to provide detailed insight into oral
conditions make this examination indispensable for planning and executing proper dental treatments
(GHODS, 2023).
Review studies have shown that the most frequent application of artificial intelligence in
dentistry is largely concentrated in the areas of radiology and oral imaging. The integration of AI in
these areas has shown significant potential to improve image interpretation, facilitating diagnoses,
providing greater speed and accuracy (CHAN, 2018).
In this sense, the primary goal of using AI in dental radiology is to increase the efficiency and
quality of services provided. AI can analyze large volumes of data faster and more accurately than
traditional methods, allowing healthcare providers to focus on more complex and personalized
aspects of patient care. In addition, it can assist in standardizing diagnoses and reducing human
errors, increasing sensitivity and promoting safer dental practice (THURZO, 2022).
Many studies are committed to improving the classification and segmentation of oral
radiographs based on machine learning. Mathematical morphology, active boundary models, level
definition methods, Fourier descriptors, textures, Bayesian techniques, linear models or binary
support vector machines are used. Classification includes a broad spectrum from the detection of the
absence or presence of a pathology or object of interest, to its classification. For the efficiency of AIs

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
in the use of images, large banks, the so-called "big data", are needed. These are increasingly robust,
especially with CT scans (GHODS, 2023).
Recently, several authors (RASTEAU, 2022; MONTERUBBIANESI, 2022; FERRO, 2019)
began to conduct studies focused on evaluating the use of AI in imaging technologies, specifically
applied to diagnostic protocols in dentistry. These studies have investigated the implementation in
several areas of oral medicine. The use of AI has been reported in a wide range of radiographic
examinations, including periapical radiographs, bite-wing (interproximal) radiographs, cone beam
computed tomography (CBCT), magnetic resonance imaging (MRI), and panoramic radiographs.
The application of AI in these exams has shown the potential to significantly improve diagnostic
accuracy, the efficiency of clinical processes, and the overall quality of dental care.
The use of AI for the analysis of CBCT and MRI has revolutionized the way facial anatomy
is evaluated and pathologies are visualized in oral and maxillofacial surgeries. These technological
advancements allow for automatic and sensitive visualization of craniofacial structures, providing a
more accurate understanding of clinical conditions and facilitating decision-making. The application
of AI is particularly effective in assessing cystic lesions and locating included third molars, allowing
surgeons to plan interventions with greater precision. The identification of osteoporosis in the jaws is
also enhanced with the use of this technology, offering valuable insights into bone density and
overall health of the jaw bones (CHOI, 2022).
In addition, AI significantly improves the detection of soft tissue calcifications and anomalies
in areas surrounding dental elements, such as mucosal thickening in the maxillary sinus and the
presence of radiopacities in the jaw. These advanced imaging capabilities assist in the accurate
diagnosis of pathological conditions, the formulation of well-informed therapeutic decisions, and the
design of detailed treatment plans (DESHMUKH, 2018).
AI also enables better prediction of health outcomes, optimizing treatment outcomes and
improving patients' quality of life. Thus, the integration of AI in the analysis of CBCT and MRI
represents a significant advance in dentistry, promoting safer, more effective, and personalized
interventions (LEE, 2022).
The integration of AI with periapical radiographic examinations, panoramic examinations,
and cone beam computed tomography scans is being widely used for the evaluation of periapical
lesions in teeth with necrotic pulp, location of the apical foramen, and the working length of the
canal. The dental surgeon, by analyzing the exams in digital format, can automatically identify
abnormal points in the region of interest and obtain the information that is necessary for the
treatment, complementing the information obtained by the anamnesis and the clinical examination
(FERRO, 2019).

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
 In oral rehabilitation, based on machine learning, it is possible to merge data from 3D images
of CT scans, intraoral scans, and facial scans, which result in the proper visualization of the patient's
morphology. The role of AI in implant dentistry is synchronized with the advancement of three-
dimensional computerized cone beam imaging and intraoral digital scans. They are useful for
methodically designing and manufacturing the implant prosthesis. Periapical and panoramic
examinations are also used to assess morphological structures in the planning of surgeries (EKERT,
2019).

APPLICATIONS IN ORTHODONTICS
Artificial intelligence (AI) is revolutionizing the paradigm of orthodontics, positioning itself
as one of the dental specialties with the highest number of collaborations with this technology.
Among all the skeletal and soft tissue structures available for orthodontic diagnosis, the face stands
out as one of the most important, although it is also the most challenging for scientific understanding
by professionals. Automated AI facial surface analysis, coming from cone beam computed
tomography (CBCT) or any other type of facial scanner, can be used to increase the accuracy and
efficiency of diagnosis at any stage of life (IMPELLIZZERI, 2020)
There are reports in the literature about the usefulness of AI-managed tools during the
COVID-19 pandemic, especially in the orthodontic retention or pre-treatment phase, allowing
clinical conditions defined by experts to be regularly assessed through patients' home videos.
AI algorithms could analyze cephalometric radiographs to automatically identify and measure
key anatomical landmarks used in orthodontic diagnosis and treatment planning. AI-powered
software assists orthodontists in creating cephalometric tracings, analyzing facial and dental
measurements, and simulating treatment outcomes based on established protocols (DHOPTE, 2023).
Recent studies have used AI to predict the need for tooth extractions based on clinical
features and cephalometric measurements. The decision between extraction and non-extraction has
been widely debated in orthodontics. In some studies (LI, 2019), AI demonstrated an 80% efficacy in
determining the best treatment approach for 200 patients with malocclusion, aged 11 to 15 years.
In addition, there are reports on the use of CAD/CAM for 3D printing of orthodontic brackets
and custom aligners. These printed devices are accompanied by algorithms that intelligently
determine how the teeth should be moved, the force to be applied, and identify specific pressure
points (SAGHIRI, 2022).
In dentofacial orthopedics, the use of Deep Learning for automated tooth segmentation in 3D
models of the jaw allows the reconstruction of teeth with their roots from CT scans. Iterative closest-
point algorithms are used to form a complete digital dental model, providing essential information
for the simulation of orthodontic treatment (BONNY, 2023).

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
APPLICATIONS IN ORAL AND MAXILLOFACIAL SURGERY AND TRAUMATOLOGY
The use of AI in Oral and Maxillofacial Surgery and Traumatology (CTBMF) has grown
significantly in recent years, helping to improve the performance of several areas of this specialty
(RASTEAU et al., 2022; ESCHERT et al., 2022). It has been used for the development of tasks
ranging from patient triage, diagnosis, treatment plan to postoperative projections (SHAN, 2021).
Among the advances provided by this technology, the identification and accuracy of
characteristic anatomical points in panoramic radiographs and cone beam computed tomography
(CBCT), through specialized software, has been one of the most significant advances in AI in
CTBMF (RASTEAU et al., 2022). The algorithms that work in these software are able to accurately
and quickly identify radiolucent bone lesions in the maxilla and mandible region.
Previous studies have reported that through models with specific image analysis algorithms,
the diagnostic accuracy made by AI was as efficient as those performed by a specialist (CHEN,
STANLEY, and ATT, 2020; RASTEAU et al., 2022). In addition, another important point is the
ability to early identify premalignant lesions in the mouth and pharynx region by means of
autofluorescence measurement and photo analysis, which correlates high-dimensional parameters
with clinical diagnoses and prognostic factors, thus constituting true radiological biomarkers
revealing characteristic points in the detection of these lesions (RASTEAU et al., 2022).
The machine learning tools involved in this system, in addition to providing the early
identification of these lesions, may be able to predict the evolution of the disease, project the
prognosis, as well as assist, through the data from electronic medical records present in their
databases, in the choice of the most appropriate treatment for each case (FONTENELE et al., 2023).
Another important advance of these tools is found in Orthognathic Surgeries. The location of
cephalometric points is a crucial factor for accurate orthodontic diagnosis and treatment procedures.
Errors in the identification of these reference points are directly linked to misdiagnosis and flawed
treatment results (VILA-BLANCO et al., 2021; FONTENELE et al., 2023;). In this context, seeking
to prevent the occurrence of these damages, automated detection of landmarks using machine
learning techniques on lateral cephalometric radiographs has emerged as a promising approach
(ESCHERT et al., 2022; FONTENELE et al., 2023).
The literature has recorded the various methods employed by authors in relation to precision
and efficiency in the detection of these points, such as the "random forest technique". However, these
more traditional spot detection techniques have limitations (MIRAGALL et al., 2023). When
compared to automated detection systems, traditional techniques are time-consuming procedures,
have a higher risk of presenting potential biases and inaccuracies associated with detected reference
point data that can be influenced by the observer's experience (MIRAGALL et al., 2023).

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
 Still dealing with aspects related to AI advances in orthognathic surgery, the possibility of
projections of the postoperative profile of patients is currently provided (ROKHSHAD, KEYHAN
and YOUSEFI, 2023). In addition, AI-aided planning based on images and 3D models, makes it
possible to simplify cephalometric analysis and operation simulation, as well as offer a more accurate
visualization of dental abnormalities, occlusal plane inclination, and body length and mandibular
branch (VILA-BLANCO et al., 2021).
The extraction of impacted third molars was also another procedure that gained prominence
with the implementation of more current technologies. This can be considered as one of the most
performed procedures in the specialty of CTBMF. The artificial neural networks that have been
introduced to optimize and guide the procedure of extracting these teeth can predict the surgical
outcome, difficulties, angulation of the tooth, as well as the relationship of the tooth to the
mandibular canal. Although neurological complications cannot be entirely avoided in cases with a
higher level of complexity, still the use of AI preoperatively significantly reduces the chances of
neurological injury. In addition, this type of technology allows a prediction for postoperative edema
complications, providing the patient with a greater level of knowledge for postoperative care
(RASTEAU et al., 2022; MIRAGALL et al., 2023).
Despite the great advance that artificial intelligence has been providing for dentistry, its
limitations in CTBMF are still strongly debated. Issues related to the limited explanation and lack of
transparency of diagnostic development provided by these technologies become factors that
corroborate the skepticism of some professionals in the field (AHMED et al., 2021). As this
technology becomes more accurate in results, its complexity also seems to evolve, limiting not only
the understanding of healthcare professionals, but also of patients (KRISHNAN, 2022; MIRAGALL
et al., 2023).
There is a study that highlights the limitation in scientific content regarding AI tools in
reputable journals. In addition, they recommend that CTBMF uniprofessional residency programs
incorporate essential AI curricula into their educational offerings to improve practitioners'
understanding of AI models and algorithms (MIRAGALL et al., 2023).
In short, although the implementation of the tools provided by AI are still recent in the field
of Oral and Maxillofacial Surgery and Traumatology, their aid in diagnosis, therapeutic decision,
preoperative planning, or prediction of the outcome of surgery is of significant importance in daily
clinical practice (RASTEAU et al., 2022; ALZAID et al., 2023). Although there are some situations
that need to be resolved in terms of ethics, data protection and diagnostic transparency, the
implementation of this technology in the dental environment can be considered as a permanent aid.

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
APPLICATIONS IN ENDODONTICS
Endodontics, a dental specialty dedicated to the diagnosis and treatment of pathologies of the
dental pulp and periapical tissues, has benefited enormously from technological advances, especially
because it is an area that requires numerous associated skills (KAROBARI et al., 2023). Among the
most promising innovations is the application of artificial intelligence, since its ability to analyze
large volumes of data offers powerful tools to improve the accuracy of diagnoses, optimize treatment
plans, and increase the efficiency of endodontic procedures (MORAES et al., 2019).
Despite advances in treatment techniques, instrumentation, and materials, epidemiological
studies reveal a high prevalence of radiographic periapical lesions and the persistence of periapical
pathology in endodontically treated teeth, which makes the use of AI relevant to assist not only in
dental clinical practice, but also to contribute to the personalization of patient care, ensuring faster,
safer, and more effective treatments, which represents a significant advance, aligning with the global
trend of incorporating cutting-edge technology to enhance patients' health care and well-being
(KAROBARI et al., 2023; PAULINO et al., 2023; SAVEGNAGO et al.,2024).
Saeed Asgary (2024), in a literature review, explored recent advances in artificial neural
networks in endodontics. This neural architecture framework comprises neurons with interconnected
robust systems that function primarily as information systems to determine problems. The results
indicated that artificial intelligence models have a great utility in analyzing the anatomy of root
canals. They allow both the early detection of periapical lesions and the precise determination of the
working length.
In addition, AI models have shown an exceptional ability to recognize landmarks and lesions
in cone-beam CT scans while maintaining consistently high accuracy rates. In addition, they have
been shown to be effective in predicting the success of treatments, identifying various conditions
such as dental caries, pulp inflammations and vertical root fractures, as well as offering second
options for non-surgical root canal treatments.
In a systematic review, Boreak (2020) pointed out that there is a range of applications of AI
implemented in endodontics, but highlighted the effectiveness of artificial intelligence applications
developed for endodontic diagnosis, decision-making support, and prognosis prediction. The study
also addressed the use of neural networks in endodontics, based mainly on convolutional neural
networks (CNNs) and artificial neural networks (ANNs). These AI systems have been applied in the
precise location of the apical foramen, prognoses related to the need for retreatment, prediction of
periapical conditions, accurate detection and diagnosis of vertical root fractures, in addition to the
thorough evaluation of root morphologies. Investigations indicate that neural networks showed
results comparable to those of experienced professionals in terms of accuracy and precision.

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
 Savegnago et al. (2024) gathered some relevant data when conducting a narrative review of
the literature. In the review, an innovative apical foramen localization system was highlighted that
achieved a 93% success rate in accurately locating the apical foramen by extracting radiographic
features and processing them in artificial neural networks. In the results regarding the detection of
periapical lesions, it was found that software using artificial intelligence presented high accuracy
rates, reaching 93% accuracy with cone beam computed tomography (CBCT) images and 85% with
panoramic radiographs. In the analysis of studies focused on root fractures, AI obtained a diagnostic
accuracy of 70% with periapical radiographs and 96% with cone beam computed tomography
(CBCT) images.
These advancements represent a remarkable opportunity for endodontics, with the potential to
radically transform the way dental professionals approach the diagnosis and treatment of endodontic
conditions. However, it is crucial that practitioners are prepared to integrate AI into their practice
ethically and responsibly, ensuring that the advantages are maximized for the benefit of patients. In
this regard, we can expect AI to continue to play an increasingly important role in the evolution of
endodontics and improving clinical outcomes for patients (KESKIN, 2021; BOREAK, 2020).

APPLICATIONS IN IMPLANTOLOGY AND PROSTHODONTICS

Implantology and Dental Prosthesis are areas of dentistry responsible for replacing lost and/or
artificially compromised dental elements. With the improvement of dental restorative materials and
AI, it is possible to deliver better results to patients and with greater efficiency in dental clinical
practice (HENDI et al., 2024).
In this context, AI has been shown to be a great ally of dentistry and especially of the
rehabilitation areas of oral health, as it plays a fundamental role in diagnosis, offering support for
treatment decisions, in addition to managing complex situations, increasing the predictability of
cases and indicating possible negative outcomes during oral rehabilitation. However, a complete
transformation from the conventional workflow to the digital workflow is required (DHOPTE et al.,
2023; FERRO et al., 2019).
From the perspective of implant dentistry, AI is commonly used to assist in diagnosis and
treatment planning, as it analyzes cone beam computed tomography images and intraoral and facial
digital scanning images of the patient, being able to simulate the installation of dental implants and
provide virtual treatment plans, displayed through three-dimensional simulations the ideal placement
position and angulation of dental implants to obtain predictable results from the biomechanical and
aesthetic point of view (FERRO et al., 2019).
The case treatment plan is carried out in software, such as DTX Studio, Blue Sky Plan,
NeoGuide System, DIOnavi and Simplant, among others, according to the information on the three-

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
dimensional anatomical and occlusal positioning of the prostheses on the implant, digitally planned
through reverse planning. Some of these software are available free of charge for digital planning
(FERRO et al., 2019).
AI helps in the identification of dental anomalies by comparing patient data and examinations
with established norms and standards, these algorithms can flag deviations and abnormalities and
notify the operator of potential problems. They collaborate in decision-making, predictability of
treatment, preoperative planning (preparation of surgical guides), recognition of implant types
through image recognition using periapical and panoramic radiographs (FERRO et al., 2019;
DHOPTE and BAGDE, 2023; TURZHO et al., 2022).
With the AI algorithm, it is possible to calculate and minimize the stress on the implant-bone
interface by observing three implant variables (the length of the implant, the length of the thread, and
the pitch of the thread), in addition to accurately calculating the modulus of elasticity of the implant-
bone interface. AI increases the predictability of osseointegration through the analysis of risk factors
and bone anatomy, along with finite element analysis calculations (HENDI et al., 2024; DHOPTE et
al., 2023; REVILLA-LEÓN et al., 2021)
In dental prosthesis, there is a total immersion of the specialty in AI, from the capture of
images of the arches via intraoral and bench scanners to the prosthetic production from milling
machines, through CAD/CAM technology. The evolution of AI in this specialty has been intense,
especially in the manufacture of removable and fixed prostheses, the highlighting of prosthetic
margins, the selection of colors, the manufacture of prostheses on implants and maxillofacial
prostheses, in addition to the establishment of a facilitated maxillo-mandibular relationship (HENDI,
et al., 2024).
The AI flow in dental prosthesis starts with the use of intraoral scanners. AI is used during the
scanning process to automatically remove excess soft tissues and materials (TURZHO et al., 2022).
In the software, the algorithm has the ability to process data and associate the relevant clinical
information of the patients entered into the system, such as fingerprints, x-rays and facial scans, to
create 3D virtual models of the patient's oral anatomy for an optimal diagnosis (DHOPTE et al.,
2023)
The software also performs the classification of dental arches in removable prosthesis, as well
as autonomously suggests different designs of structures considered more appropriate for the clinical
case, assists in the design of the prosthesis, in the determination of its type (removable or fixed) and
in the selection of components, identify periodontally compromised teeth, in addition to delimiting
the areas of end of the margin of preparations with onlay ceramic purposes, fixed inlays, crowns and
bridges requiring only minor adjustments at the operator's expense, improving the accuracy of final

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
dentures and decreasing the need for manual adjustments (THURZO et al., 2022; FERRO et al.,
2019; DHOPTE et al., 2023)
The data generated by three-dimensional machines is used by AI to design the prosthesis on
implant in reverse planning, thus determining the most convenient position and angulation for
implantation and fabrication of precise surgical guides (AL HENDI, et al., 2024).
In conventional prosthetics, the integration of AI has increased accuracy, precision, and
reliability, positively impacting clinical outcomes. During digital design, the CAD/CAM tool is used
for the manufacture of fixed and removable dental restorations. In this process, a virtual library with
millions of natural crowns can be used to create the best possible crown design for the various
clinical situations, in addition to the software making it possible to recreate acceptable intermaxillary
relationships, configuring the correctly designed teeth, evaluating occlusal relationships and
identifying interferences in search of the appropriate functional dynamics. Virtual analysis of the
joint and occlusion can assist in the planning of prosthetic treatment, ensuring proper occlusal
alignment and harmony for appropriate functional and aesthetic outcomes (DHOPTE et al., 2023;
TURZHO et al., 2022).
The models made in the software serve as a reliable diagnostic tool for selecting the color of
the tooth, presenting varied shapes and enabling adjustments in aesthetics in search of the best results
for the patient during the digital design of the smile, in addition to predicting facial changes in
patients with removable dentures and design of removable partial dentures. The software is also able
to predict the color change of prosthetic restorations in various lighting conditions (DHOPTE et al.,
2023; BONNY et al., 2023).
In addition to the models, smile design uses 3D facial scanning and virtual resources along
with segmented computed tomography, intraoral scans, and facial scans, resulting in the
virtualization of the patient's morphology, producing a facial design with the patient's smile
(TURZHO et al., 2022).
CAD/CAM software not only designs, but is also responsible for manufacturing, printing, or
milling the prosthesis. Commonly used to manufacture inlays, onlays, crowns and fixed bridges.
Resulting in savings in time, resources, and energy for the dentist, prosthetist, and patients, in
addition to reducing the chances of human error in the final prosthesis and performing unnecessary
procedures (HENDI, et al., 2024).

APPLICATIONS IN FORENSIC DENTISTRY

Forensic Dentistry, also called Forensic Dentistry, is a specialty of dentistry that employs
technical-scientific processes to assist in human identification in forensics focused on the head and
neck. This includes the analysis of dental arches, the comparison of dental records, facial

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
reconstruction, and the identification of other anatomical structures in the region, making a crucial
contribution to legal and judicial investigations (RAMOS et al., 2021).
The importance of the forensic dentist has become crucial in forensic science, particularly
due to the histological characteristics of teeth that make them resistant, such as not sustaining
damage when subjected to temperatures of 600°C and remaining intact for long periods after the
body has deteriorated. Thus, when a body is found in conditions such as skeletonization, charring,
advanced stages of decomposition, or other situations that make identification difficult or impossible
by traditional methods, such as dactyloscopy, the analysis of the tooth structure becomes
fundamental for individual identification (BIANCH, 2019).
Advances in computer science and radiology refinement have played a significant role in
achieving greater accuracy and detailed image analysis, providing enlargements without loss of
quality, reducing image processing time, and offering various adjustment options. These advances
have enabled more accurate identifications and facilitated the comparison between ante-mortem and
post-mortem radiographic examinations (RAMOS et al., 2021; BIANCH, 2019).
Nascimento Neto et al. (2019) conducted a research in order to map and categorize forensic
knowledge, with the purpose of developing a methodological plan to support the creation of a tool
based on artificial intelligence specific to Forensic Dentistry. The study presented a proposal that
uses artificial intelligence to store dental records, perform comparisons and recognize radiographic
images, ensuring the legitimacy and authenticity of the data.
The software developed in the study had the participation of the dental surgeon, the radiology
companies and the forensic dentist, and consists of a graphical interface that acts online and allows
the user (dental surgeon) to enter the dental documentation register, consisting of medical records,
radiographs, odontograms and photographs. In this way, it is possible to create a database with
accurate information on an individual, which allows the identification of the level of similarity
between ante-mortem and post-mortem examinations and, based on this, to be able to give a report
confirming or not the identification of the victim, which helps and optimizes the work of the forensic
dentist (NASCIMENTO NETO et al., 2019; BIANCH, 2019).
At the same time, (MOLLO FILHO; MELANI, 2022), demonstrated through a literature
review the relevance of the use of intraoral scanners and their applicability in the practice of forensic
dentistry. The resource is capable of generating three-dimensional digital models of the oral cavity
that can be stored, analyzed, and compared using proprietary applications.
With the application of the technique, records are obtained with agility and precision and the
data can be obtained, if necessary, in the very place where the body is located during the expert
evaluation. Research shows that intraoral scanners have as a differential the possibility of

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
differentiating monozygotic twins by analyzing palatine roughness, an anatomical detail that is also
capable of individualizing each person (GIOSTER-RAMOS et al., 2021; Bianch, 2019).
Thus, artificial intelligence combined with imaging exams and comparison algorithms have
proven to be extremely important resources in the human identification process.

PERSPECTIVES OF AI IN DENTISTRY
With the advent and applicability of AIs in healthcare, the possibility of virtual care through
teledentistry has been highlighted as promising and has the potential to reach patients in more remote
regions with limited access to dental care, so that they can, from this new resource, receive
consultations more quickly and efficiently (AGRAWAL and NIKHADE, 2022; GHODS et al.,
2023).
Among the branches of dentistry with the greatest applicability of these technological
resources, without a doubt, diagnosis and treatment have been the most impacted. Machine learning
algorithms are capable of analyzing radiographic images, CT scans, and other imaging tests,
providing faster and more accurate identification of conditions such as caries, fractures, periodontal
diseases, and even premalignant lesions, which has been helping in early diagnosis and may soon be
considered an essential tool for clinical use to minimize the chances of a poor prognosis for the
patient (CHWENDICKE, SAMEK and KROIS, 2020).
In addition, software that provides augmented reality and 3D images reveal that AI is a
promising tool for the evolution of more effective dental diagnoses (PETHANI, 2021; BONNY et
al., 2023). A study demonstrated the efficiency of an artificial neural network in providing a
diagnosis with good accuracy. In this study, a comparison was made between the diagnosis made by
a specialist and by ANN, the result showed that AI had a high sensitivity and specificity in the
diagnosis presented (TANDON, RAJAWAT and BANERJEE, 2020).
With regard to the individualization of treatments, advanced algorithms are able to analyze
previously provided data on each patient's oral health and medical-dental history, allowing, which
will allow dentists to create personalized treatment plans in daily clinical practice (SHAN, 2021).
Regarding monitoring and prevention, the literature reports that through RNA it was possible
to create a model capable of predicting the possibility of toothache in patients. This study used the
association between toothache, frequency of brushing, flossing, and other factors such as diet and
physical activity. The model made in this study was able to recognize proper eating habits, oral
hygiene, and stress prevention as the most important factors in the prevention of dentistry
(TANDON, RAJAWAT, and BANERJEE, 2020).
In addition to the above-mentioned points, advances in surgical planning have shown promise
for application in the dental environment (ALZAID et al., 2023). Through software specialized in

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
machine learning , it is possible to perform complex surgeries, such as implant placement or
orthognathic surgeries, with greater precision. This software aims to make procedures less invasive,
which will reduce recovery time and provide greater safety for the patient (RODRIGUES, KROIS
and SCHWENDICKE, 2021; ALZAID et al., 2023).
Despite the promising results that artificial intelligence provides for the dental environment,
several authors still debate the difficulties and flaws presented by this technology. Among the main
issues raised are the high complexity of understanding these systems and the high number of health
data used in previous tests for the development of these systems, which are at risk of leakage
(CHWENDICKE; SAMEK; KROIS, 2020).
Some authors also highlight the automation bias that occurs when some professionals
underestimate the possibility of error in the information presented by AI (GLICK et al., 2022). The
fact that the information was generated by a computer is enough for the professional to adopt the
result as an absolute truth, even if he sometimes has doubts about the result, this prevents him from
investigating further, allowing the chance of error (GLICK et al., 2022).

LEGAL ASPECTS OF AI IN DENTISTRY

Currently, there are no laws specifically set for the use of AI in dentistry. (ROGANOVIĆ and
RADENKOVIĆ, 2023; ROGANOVIĆ, RADENKOVIĆ and MILIČIĆ, 2023; ROKHSHAD et al.,
2023).
Medical liability currently defined under the law is inadequate and therefore may not
encourage the safe use of AI in clinical decision-making. It can be stated with certainty that the use
of AI in clinical decision-making needs dental surveillance, and the role of dentists is crucial in
preventing dental complications as well as reviewing AI systems (ROGANOVIĆ and
RADENKOVIĆ, 2023).
To respect the interaction between man and technology in a clinical setting, AI in medicine
and dentistry must have a complementary role in the work of clinical professionals. In dentistry,
various software-type algorithms are used as the basic application of AI, which is supposed to
improve the accuracy of dental diagnosis, provide visualization of anatomical guidelines during
treatment, and, due to the possibility of analyzing large amounts of data, predict the occurrence and
prognosis of oral diseases. The conscious and ethical use of AI in dentistry should consider: (1) when
to apply AI and (2) how to use AI appropriately and responsibly. Patients should be notified about
how their data is used, also about the involvement of AI-based decision-making, especially if there is
a lack of regulatory policy if AI is used to decrease costs rather than improve patients' health or if the
dentist has a conflict of interest. As many dentists are accelerating in the direction of integrating AI

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
systems into dental diagnostics, prognosis, and treatment, the legal and ethical issues are becoming
even more pertinent. (ROGANOVIĆ, RADENKOVIĆ and MILIČIĆ, 2023).
Notably, particularly for AI, considering a network of multiple stakeholders, including
patients, clinicians, software developers, and society, as well as the specific requirements, needs, and
constraints of the target medical specialty may be relevant; existing guidelines applied to medicine
will not be entirely applicable to dentistry. A specific framework and guideline on dental AI is
needed, which should facilitate dissemination to the dental domain and help optimize dental AI in
relation to its ethical foundations (ROKHSHAD et al., 2023).

FINAL THOUGHTS
AI has become an excellent tool for dentistry, providing several advantages that have
improved both clinical practice and the patient experience. The potential of AI to transform the way
dentists deal with the diagnosis, planning, and treatment of their patients is remarkable.
Although there are limitations and flaws presented by AI resources that still need to be
addressed, the advantages presented by these tools stand out and show promise in the dental field.
Still, the employment of these technologies in clinical dental practice is a challenge and the lack of
acceptance of these tools can be attributed to the complexity of dealing with new technologies and
the difficulty of acquiring new skills.
All stakeholders in this process, such as dentists, patients, and system developers, should
consider and prioritize, among other factors, well-being, privacy, the law, and respect for autonomy
and decision-making when developing, implementing, or receiving dental AI.

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
 REFERENCES

1. Ahmed, N., et al. (2021). Artificial Intelligence Techniques: Analysis, Application, and Outcome in
Dentistry—A Systematic Review. *BioMed Research International*, 2021, e9751564.
https://doi.org/10.1155/2021/9751564. Acesso em: 25/04/2024.

2. Al Hendi, K. D., et al. (2024). Artificial intelligence in prosthodontics. *Bioinformation*, 20(3),

238–242. https://doi.org/10.6026/97320630020242. Acesso em: 25/04/2024.

3. Song, Y. B., et al. (2022). Comparison of detection performance of soft tissue calcifications using
artificial intelligence in panoramic radiography. *Scientific Reports*, 12(1), 19115.
https://doi.org/10.1038/s41598-022-23963-9. Acesso em: 25/04/2024.

4. Lee, S., Kim, D., & Jeong, H. G. (2022). Detecting 17 fine-grained dental anomalies from panoramic
dental radiography using artificial intelligence. *Scientific Reports*, 12(1).
https://doi.org/10.1038/s41598-022-12685-9. Acesso em: 25/04/2024.

5. Ghods, K., et al. (2023). Application of Artificial Intelligence in Clinical Dentistry, a Comprehensive
Review of Literature: Machine learning; Deep learning; Diagnostic System. *J Dent Shiraz Univ
Med Sci*, 24(4). https://doi.org/10.34172/jodsd.2023.4. Acesso em: 25/04/2024.

6. Quintero-Rojas, J., & González, J. D. (2021). Use of Convolutional Neural Networks in

Smartphones for the Identification of Oral Diseases Using a Small Dataset. *Revista Facultad de
Ingeniería*, 30(55), e11846. https://doi.org/10.17533/udea.redin.20210425. Acesso em:
25/04/2024.

7. Dhopte, A., et al. (2023). Smart Smile: Revolutionizing Dentistry With Artificial Intelligence.
*Cureus*, 15(6). https://doi.org/10.7759/cureus.16234. Acesso em: 25/04/2024.

8. Wang, Y., et al. (2020). Developing Children’s Oral Health Assessment Toolkits Using Machine
Learning Algorithm. *JDR Clinical and Translational Research*, 5(3), 233–243.
https://doi.org/10.1177/2380084419868505. Acesso em: 25/04/2024.

9. Carvalho, D. K. de, et al. (2021). Benefício da inteligência artificial para o diagnóstico precoce da
cárie dentária: Revisão integrativa. *Research, Society and Development*, 10(4),
e43210413083. https://doi.org/10.33448/rsd-v10i4.13083. Acesso em: 25/04/2024.

10. Devlin, H., et al. (2021). The ADEPT study: a comparative study of dentists’ ability to detect
enamel-only proximal caries in bitewing radiographs with and without the use of AssistDent
artificial intelligence software. *British Dental Journal*, 231(8), 481–485.
https://doi.org/10.1038/s41415-021-3335-6. Acesso em: 25/04/2024.

11. Al-Namankany, A. (2023). Influence of Artificial Intelligence-Driven Diagnostic Tools on

Treatment Decision-Making in Early Childhood Caries: A Systematic Review of Accuracy and
Clinical Outcomes. *Dentistry Journal*, 11(9), 214. https://doi.org/10.3390/dj11090214. Acesso
em: 25/04/2024.

12. Tandon, D., & Rajawat, J. (2020). Present and future of artificial intelligence in dentistry. *Journal
of Oral Biology and Craniofacial Research*, 10(4), 391–396.
https://doi.org/10.1016/j.jobcr.2020.09.015. Acesso em: 25/04/2024.

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
13. Kühnisch, J., et al. (2021). Caries Detection on Intraoral Images Using Artificial Intelligence.
*Journal of Dental Research*, 101(2), 158–165. https://doi.org/10.1177/00220345211038665.
Acesso em: 25/04/2024.

14. Gómez, S., et al. (2022). Pasado, presente y futuro de la cariología. *International Journal of
Interdisciplinary Dentistry*, 15(3), 250–254. https://doi.org/10.2139/ssrn.3677825. Acesso em:
25/04/2024.

15. Cabral, B. M. de S., et al. (2021). Benefícios da inteligência artificial na identificação de cárie
dentária: revisão integrativa. *Research, Society and Development*, 10(2), e18310212117.
https://doi.org/10.33448/rsd-v10i2.12117. Acesso em: 25/04/2024.

16. Pethani, F. (2021). Promises and perils of artificial intelligence in dentistry. *Australian Dental
Journal*, 66(1), 9–18. https://doi.org/10.1111/adj.12814. Acesso em: 25/04/2024.

17. Arsiwala-Scheppach, L. T., et al. (2024). Impact of artificial intelligence on dentists’ gaze during
caries detection: A randomized controlled trial. *Journal of Dentistry*, 140, 104793.
https://doi.org/10.1016/j.jdent.2023.104793. Acesso em: 25/04/2024.

18. Chaves, E. T., et al. (2024). Detecção de Cárie ao Redor de Restaurações em Mordidas Usando
Aprendizado Profundo. *Revista de Odontologia*, 143, 104886–104886.
https://doi.org/10.29327/14871. Acesso em: 25/04/2024.

19. Chan, M., et al. (2018). Accuracy of extraoral bite-wing radiography in detecting proximal caries
and crestal bone loss. *The Journal of the American Dental Association*, 149(1), 51–58.
https://doi.org/10.1016/j.adaj.2017.10.022. Acesso em: 25/04/2024.

20. Thurzo, A., et al. (2022). Where Is the Artificial Intelligence Applied in Dentistry? Systematic
Review and Literature Analysis. *Healthcare (Basel, Switzerland)*, 10(7), 1269.
https://doi.org/10.3390/healthcare10071269. Acesso em: 25/04/2024.

21. Chen, H., et al. (2019). A deep learning approach to automatic teeth detection and numbering based
on object detection in dental periapical films. *Scientific Reports*, 9(1), 3840.
https://doi.org/10.1038/s41598-019-40347-7. Acesso em: 25/04/2024.

22. Rasteau, S., et al. (2022). Artificial intelligence for oral and maxillo-facial surgery: A narrative
review. *Journal of Stomatology, Oral and Maxillofacial Surgery*, 123(3), 276–282.
https://doi.org/10.1016/j.jormas.2021.12.002. Acesso em: 25/04/2024.

23. Monterubbianesi, R., et al. (2022). Augmented, Virtual and Mixed Reality in Dentistry: A Narrative
Review on the Existing Platforms and Future Challenges. *Applied Sciences*, 12(2), 877.
https://doi.org/10.3390/app12020877. Acesso em: 25/04/2024.

24. Ferro, V., Nicholson, J., & Koka, S. (2019). Innovative Trends in Implant Dentistry Training and
Education: A Narrative Review. *Journal of Clinical Medicine*, 8(10), 1618.
https://doi.org/10.3390/jcm8101618. Acesso em: 25/04/2024.

25. Lee, S., Kim, D., & Jeong, H. G. (2022). Detecting 17 fine-grained dental anomalies from
panoramic dental radiography using artificial intelligence. *Scientific Reports*, 12(1).
https://doi.org/10.1038/s41598-022-12685-9. Acesso em: 25/04/2024.

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
26. Ekert, T., et al. (2019). Deep Learning for the Radiographic Detection of Apical Lesions. *Journal
of Endodontics*, 45(7), 917-922.e5. https://doi.org/10.1016/j.joen.2019.04.002. Acesso em:
25/04/2024.

27. Keskin, C., & Keleş, A. (2021). Digital Applications in Endodontics. *Journal of Experimental and
Clinical Medicine*, 38(SI-2), 168–174. https://doi.org/10.52142/omujecm.v38iSI.2.01. Acesso
em: 25/04/2024.

28. Lee, S., & Kim, J.-E. (2022). Evaluating the Precision of Automatic Segmentation of Teeth,
Gingiva and Facial Landmarks for 2D Digital Smile Design Using Real-Time Instance
Segmentation Network. *Journal of Clinical Medicine*, 11(3), 852.
https://doi.org/10.3390/jcm11030852. Acesso em: 25/04/2024.

29. Alzaid, N., et al. (2023). Revolutionizing Dental Care: A Comprehensive Review of Artificial
Intelligence Applications Among Various Dental Specialties. *Cureus*, 15(10).
https://doi.org/10.7759/cureus.16242. Acesso em: 25/04/2024.

30. Chen, Y.-W., Stanley, K., & Att, W. (2020). Artificial intelligence in dentistry: current applications
and future perspectives. *Quintessence International (Berlin, Germany: 1985)*, 51(3), 248–257.
https://doi.org/10.3290/j.qi.a43904. Acesso em: 25/04/2024.

31. Eschert, T., et al. (2022). A Survey on the Use of Artificial Intelligence by Clinicians in Dentistry
and Oral and Maxillofacial Surgery. *Medicina*, 58(8), 1059.
https://doi.org/10.3390/medicina58081059. Acesso em: 25/04/2024.

32. Fatima, A., et al. (2022). Advancements in Dentistry with Artificial Intelligence: Current Clinical
Applications and Future Perspectives. *Healthcare*, 10(11), 2188.
https://doi.org/10.3390/healthcare10112188. Acesso em: 25/04/2024.

33. Fontenele, R. C., et al. (2023). Convolutional neural network-based automated maxillary alveolar
bone segmentation on cone-beam computed tomography images. *Clinical Oral Implants
Research*, 34(6), 565–574. https://doi.org/10.1111/clr.13962. Acesso em: 25/04/2024.

34. Krishnan, D. G. (2022). Artificial Intelligence in Oral and Maxillofacial Surgery Education. *Oral
and Maxillofacial Surgery Clinics of North America*, 34(4), 585–591.
https://doi.org/10.1016/j.coms.2022.06.009. Acesso em: 25/04/2024.

35. Miragall, M. F., et al. (2023). Face the Future—Artificial Intelligence in Oral and Maxillofacial
Surgery. *Journal of Clinical Medicine*, 12(21), 6843. https://doi.org/10.3390/jcm12216843.
Acesso em: 25/04/2024.

36. Rokhshad, R., Keyhan, S. O., & Yousefi, P. (2023). Artificial intelligence applications and ethical
challenges in oral and maxillo-facial cosmetic surgery: a narrative review. *Maxillofacial Plastic
and Reconstructive Surgery*, 45(1). https://doi.org/10.1186/s40902-023-00345-7. Acesso em:
25/04/2024.

37. Shan, T., Tay, F. R., & Gu, L. (2021). Application of Artificial Intelligence in Dentistry. *Journal
of Dental Research*, 100(3), 232–244. https://doi.org/10.1177/0022034520969115. Acesso em:
25/04/2024.

38. Vila-Blanco, N., et al. (2021). Automated description of the mandible shape by deep learning.
*International Journal of Computer Assisted Radiology and Surgery*, 16(12), 2215–2224.
https://doi.org/10.1007/s11548-021-02421-1.

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
39. Agrawal, P., & Nikhade, P. (2022). Artificial Intelligence in Dentistry: Past, Present, and Future.
*Cureus*, 14(7). https://doi.org/10.7759/cureus.27360.

40. Alzaid, N., et al. (2023). Revolutionizing Dental Care: A Comprehensive Review of Artificial
Intelligence Applications Among Various Dental Specialties. *Cureus*, 15(10).
https://doi.org/10.7759/cureus.37850.

41. Bonny, T., et al. (2023). Contemporary Role and Applications of Artificial Intelligence in Dentistry.
*F1000Research*, 12, 1179–1179. https://doi.org/10.12688/f1000research.11709.1.

42. Ghods, K., et al. (2023). Application of Artificial Intelligence in Clinical Dentistry, a
Comprehensive Review of Literature Machine learning; Deep learning; Diagnostic System. *J
Dent Shiraz Univ Med Sci*, 24(4). https://doi.org/10.30476/dentjods.2023.99316.

43. Glick, A., et al. (2022). Impact of explainable artificial intelligence assistance on clinical decision-
making of novice dental clinicians. *JAMIA Open*, 5(2).
https://doi.org/10.1093/jamiaopen/ooac001.

44. Pethani, F. (2021). Promises and perils of artificial intelligence in dentistry. *Australian Dental
Journal*. https://doi.org/10.1111/adj.12806.

45. Rodrigues, J. A., Krois, J., & Schwendicke, F. (2021). Demystifying artificial intelligence and deep
learning in dentistry. *Brazilian Oral Research*, 35. https://doi.org/10.1590/1807-3107bor-
2021.vol35.0075.

46. Schwendicke, F., Samek, W., & Krois, J. (2020). Artificial Intelligence in Dentistry: Chances and
Challenges. *Journal of Dental Research*, 99(7), 769–774.
https://doi.org/10.1177/0022034520915714.

47. Shan, T., Tay, F. R., & Gu, L. (2021). Application of Artificial Intelligence in Dentistry. *Journal
of Dental Research*, 100(3), 232–244. https://doi.org/10.1177/0022034520969115.

48. Thurzo, A., Kurilová, V., & Varga, I. (2021). Artificial Intelligence in Orthodontic Smart
Application for Treatment Coaching and Its Impact on Clinical Performance of Patients
Monitored with AI-TeleHealth System. *Healthcare*, 9(12), 1695.
https://doi.org/10.3390/healthcare9121695.

49. Impellizzeri, A., et al. (2020). Dental Monitoring Application: it is a valid innovation in the
Orthodontics Practice? *La Clinica Terapeutica*, 171(3), e260–e267.
https://doi.org/10.7417/CT.2020.2222.

50. Dhopte, A., et al. (2023). Smart Smile: Revolutionizing Dentistry With Artificial Intelligence.
*Cureus*, 15(6). https://doi.org/10.7759/cureus.37826.

51. Li, P., et al. (2019). Orthodontic Treatment Planning based on Artificial Neural Networks.
*Scientific Reports*, 9(1). https://doi.org/10.1038/s41598-019-40286-3.

52. Revilla-León, M., et al. (2021). Artificial intelligence applications in implant dentistry: A
systematic review. *The Journal of Prosthetic Dentistry*.
https://doi.org/10.1016/j.prosdent.2021.02.005.

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice
53. Data Science Academy. (2022). *Deep Learning Book*. Disponível em:
<https://www.deeplearningbook.com.br/>. Acesso em: 10 junho. 2024.

54. Gioster-Ramos, M. L., et al. (2021). Técnicas de identificação humana em Odontologia Legal.
*Research, Society and Development*, 10(3), e20310313200–e20310313200.
https://doi.org/10.33448/rsd-v10i3.13200.

55. Nascimento Neto, C. D., et al. (2019). Inteligência artificial como ferramenta para identificação
humana em odontologia legal. *Brazilian Journal of Production Engineering*, 5(4), 82-96.
https://doi.org/10.47244/bjpe.2019.v5.n4.07.

56. Bianch B. V. (2019). A Importância da Atuação do Odontolegista na Identificação Pessoal de

Corpos Carbonizados e a Determinação do Gênero. *Revista UNINGÁ*, 56(S3), 119-129.

57. Mollo Filho, P. C., & Melani, R. F. H. (2022). UTILIZAÇÃO DE ESCÂNERES INTRAORAIS
NA ODONTOLOGIA LEGAL: UMA REVISÃO DE LITERATURA. *Revista Brasileira de
Odontologia Legal*, 89–97. https://doi.org/10.4013/rbol2022.321.12.

58. Boreak, N. (2020). Effectiveness of Artificial Intelligence Applications Designed for Endodontic
Diagnosis, Decision-making, and Prediction of Prognosis: A Systematic Review. *The Journal
of Contemporary Dental Practice*, 21(8), 926–934. https://doi.org/10.5005/jp-journals-10024-
2792.

59. Asgary, S. (2024). Artificial Intelligence in Endodontics: A Scoping Review. *PubMed*, 19(2),
85–98. https://doi.org/10.1177/00084098211063505.

60. Karobari, M. I., et al. (2023). Evaluation of the Diagnostic and Prognostic Accuracy of Artificial
Intelligence in Endodontic Dentistry: A Comprehensive Review of Literature. *Computational
and Mathematical Methods in Medicine*, 2023, 7049360.
https://doi.org/10.1155/2023/7049360.

61. Paulino, G. L., et al. (2023). Como a inteligência artificial pode ser utilizada na endodontia?
*Revista do CROMG*, 22(Supl.4).

62. Savegnago, G. D. O., et al. (2022). Inteligência artificial na odontologia: uma revisão narrativa de
literatura. *Repositório da Faculdade de Odontologia de Passo Fundo Federal*, 29(1), 1.

63. Deshmukh, S. (2018). Artificial intelligence in dentistry. *Journal of the International Clinical
Dental Research Organization*, 10(2), 47. https://doi.org/10.4103/jicdro.jicdro_21_18.

64. Roganović, J., Radenković, M., & Miličić, B. (2023). Responsible Use of Artificial Intelligence in
Dentistry: Survey on Dentists’ and Final-Year Undergraduates’ Perspectives. *Healthcare*,
11(10), 1480. https://doi.org/10.3390/healthcare11101480.

65. Roganović, J., Radenković, M., & Ethical use of IA in Dentistry. (2023). *IntechOpen eBooks*.

66. Rokhshad, R., et al. (2023). Ethical considerations on artificial intelligence in dentistry: A
framework and checklist. *Journal of Dentistry*, 135, 104593.
https://doi.org/10.1016/j.jdent.2023.104593.

Challenges and Research in Health Sciences: A Multidisciplinary Approach

Artificial intelligence and dental clinical practice