Artificial intelligence (AI) is a machine-based system that can make predictions and recommendations based on definitive objectives given to the system by humans. One of the subfields of AI is machine learning (ML). This technique involves algorithms and statistics. This is used to predict unknown data. Further, the interesting aspect of ML is that, the ability to self-learn and improve from its own experience without being explicitly programmed. Several proprietary algorithms are employed, and a complete description of them is beyond the purview of this review. Neural networks (NNs) are established machine learning models that can be applied to complicated data structures. Artificial neurons (ANs) form the basis of a neural network. These ANs are akin to human neurons in function. Furthermore, several layers of neural networks constitute Deep learning (DL) and it is beneficial for managing sophisticated data. convolutional neural networks (CNNs) are DL algorithms, widely used in dentistry due to their utility in diagnosis and detection. ^, “Training” and “testing” are the dual stages of AI’s functioning. The training data defines the frameworks for the model set. The model utilizes the data from preceding instances, such as from datasets and patient data. Subsequently, these settings are implemented in the test sets.

As alluded to earlier, AI is a relatively new science. A brief history of the evolution of AI is represented in Fig. 1 .

Timeline of the evolution of artificial intelligence. ^,,,,,

AI is already used in multiple health care settings, science, and technology. During the COVID-19 pandemic, AI was utilized to predict the anticipated course of the disease, based on data from imaging and laboratory tests. AI-integrated wearable health technology is employed in devices that monitor patient variables, such as vital signs, to alert them about a possible medical emergency. AI helps optimize infrastructure design, traffic system operation, and forecast structural performance. AI is used to postulate, create study designs, and gather and analyse data. AI has also been instrumental in predicting three-dimensional protein structures.

The term “diagnostic cockpit” was introduced recently to designate the integration of imaging and clinical data for early and accurate diagnoses. AI personalizes care by analyzing genetics, lifestyle, history, tumor detection, and treatment outcome prediction. AI hastens the process of drug discovery through computational drug design. AI-powered robotics improves surgeries, such as robot-assisted nephrectomies. AI automates medical documentation, billing, and scheduling, ^, offers evidence-based treatment guidance, risk prediction, and expands telemedicine, Internet of Things (IoT)-enabled remote care and mental health support.

AI is poised to revolutionize the field of dentistry. AI is transforming data integration by combining medical and dental histories, imaging, clinical records, and even social media data by analyzing this broad spectrum of information. In clinical dental practice, AI is used in various ways, from detecting cavities through image analysis to utilizing virtual reality for patient education, training, and streamlining electronic record management. The potential applications of AI models in dentistry are summarized in Fig. 2 ; AI workflow in diagnosing and treatment planning in dentistry is depicted in Fig. 3 .

Artificial intelligence models and their potential applications in dentistry. ^,,,

Artificial intelligence workflow in the diagnosis and treatment planning in dentistry. ^,

AI is promising to have a high impact on dental imaging.

AI has been utilized to compute the prevalence of periodontitis in the world population.

AI is revolutionizing prosthodontic practices by improving treatment strategies and advancing diagnosis, which may enhance treatment outcomes. The similarity is further explained in Fig. 4 .

Clinical applications of artificial intelligence in diagnosis and prosthetic production.

The accuracy of the implant outcome prediction using AI models with radiographic and diagnostic data ranged from approximately 70% to 95%. AI’s efficiency in forecasting implant outcomes depends upon bone levels around the implants, implant retention, crestal bone loss, failure of dental implants, and their success, and bone and implant integration. AI helps to analyze the implant position, angle, and depth. However, no meaningful difference was observed between the bone levels measured manually and the AI models for diagnosing peri-implantitis. The precision for the initial anchorage of implants based on the implant insertion protocols using AI technology was measured up to approximately 94%.

AI has been proposed to augment the clinical procedures in the OMFS and to improve precision. Fig. 5 summarizes the potential of AI in OMFS.

Integration of artificial intelligence to enhance various aspects of oral and maxillofacial surgery.

AI may provide a fast, precise, and reliable measurement of molar angulations, aiding in extractions. ^,, AI is used for 3D imaging techniques that refines surgical simulation and simplifies surgical planning with enhanced visualization. AI can examine preoperatively using computed tomography (CT) scans and 3D models.

The potential applications of AI in endodontics range from caries detection to predicting retreatment and assessing pulpal stem cell viability. A summary of these applications is given in Fig. 6 .

Potential applications of artificial intelligence in endodontics. ^,,,,,,,, (Illustration by Dr. Gifty Francis Ruby, BDS.). DD, differential diagnosis; EPT, electrical pulp test; P, periapical lesions; RC, root canal; VRF, vertical root fracture.

AI drastically reduced the time required to render a histopathologic diagnosis compared to human investigator’s system. A DL model was also successfully employed to predict malignant transitions. Similar applications have also been used in HPV infections, tumor histopathology, and in vivo optical biopsy principles. ^,, In addition, AI models have also been applied to dysplastic features of oral mucosal and hard tissue lesions. ^, AI may also assist in the differential diagnosis of cysts and oral white lesions, potentially negating the need for a biopsy. ^,, AI can increase the overall precision of test analysis, reduce errors in diagnosis, and enhance the efficacy of oral pathology laboratories.

AI plays a role in orthodontics by advancing technology that may help predict outcomes and plan treatment. Fig. 7 summarizes the potential applications of AI in orthodontics.

Overview of artificial intelligence in the field of orthodontics.

Identifying the reference points on the lateral cephalograms aids with better diagnosis, treatment planning, and treatment options. Manual detection of these points can be laborious, and skill based. AI-aided cephalometric assessment seems promising in this regard, combined with direct human supervision. AI may be a relatively user-friendly technology with applications in orthodontics. AI can be used to predict the need for extractions in orthodontic cases where space analysis is crucial, and a large amount of data considering a diverse population was used to indicate the extraction outcome with an efficiency of approximately 93% and also for growth and cervical maturation value predictions. AI can distinguish permanent from primary teeth, and when combined with cone-beam computed tomography (CBCT) evaluation, render 360-degree visualization of dental and maxillofacial structures, thereby enhancing orthodontic efficiency. AI may help flag high-risk areas and assist in selecting sites for implant placements for reinforced anchorage. ^, However, more research advances and improvements are needed before AI can reliably deliver computer-aided treatment plans.

AI-driven software can help dentists replicate the results of aesthetic dentistry procedures. AI tools can facilitate virtual treatment plans that adhere to accepted dental principles. AI may help measure aesthetic outcomes of orthognathic surgery cases by analyzing facial attractiveness, and age predictions. DL has shown superiority in predicting postoperative soft tissue profiles in mandibular advancement surgery. AI models have been employed to improve the accuracy of dental shade matching and can estimate changes produced by professional whitening treatments. ^, In digital smile designing, AI can help label and merge data from facial and intraoral images, potentially making the formulation of treatment plans easier. These algorithms may help generate customized 3D smile designs, possibly minimizing the need for mock-ups and conventional wax-ups. ^,

AI revolutionizes forensic dentistry by enhancing identification, age estimation, and facial reconstruction using image processing and pattern recognition. It supports bite mark analysis, disaster victim identification, and fraud detection via dental record analysis. Radiographs aid facial identification by comparing antemortem and postmortem data, with new algorithms extracting cranial patterns. AI may predict gender from radiographs based on specific landmarks. AI may help forensic identification by analyzing sexually dimorphic parameters of the skull and pelvis.

AI models can aid in life-saving diagnosis of facial pain and temporomandibular disorders, along with neuropathic, neurovascular, and cardiac referred pain. ^, AI helps nonspecialists assess headaches, their subtypes, and pain patterns by employing a questionnaire. AI models aid in diagnosing epilepsy, movement disorders, and neurocognitive conditions. AI successfully interprets trigeminal neuralgia by understanding the specific features of cortical and subcortical regions, demarcating the cerebellopontine angle, and differentiating myofascial pain. AI may also create a postsurgical prognostic framework to predict orofacial pain management.

AI may enhance health protection, promotion, and support public health surveillance in community dentistry. AI models may predict and enhance risk assessments, timely recognition, diagnostic predilection, and customized treatment plans for cancer in multiple organs. AI along with radiographic and biomarker data, could be used to detect to identify high-risk oral disease populations for timely interventions. AI is helpful in microbiome management through polymer and peptide synthesis and in the development of vaccines.

The World Health Organization and the International Telecommunication Union launched initiatives to standardize AI in health care by developing guidelines on ethics, regulation, technology, and clinical use, including dental-specific tools like checklists and training programs. The American Dental Association also released guidelines detailing image analysis standards, highlighting the importance of continuous monitoring, human oversight, and cybersecurity. Strict protocols must be followed to confirm the clinical effectiveness and proper supervision of autonomous AI tools. Before implementing AI in clinical settings, it’s essential to ensure transparency, reliability, and data security.