Additive manufacturing is a rapid production process of any three-dimensional (3D) object using 3D printers. The process allows a complex geometrical design to be produced with additional benefits such as a reduction of unnecessary raw material wastage, mass production of the desired items, and fast production and manufacturing of dental prostheses as compared to subtractive manufacturing [31]. In dentistry, additive manufacturing has been adopted in multiple dental fields such as prosthodontics, implant dentistry, oral surgery, and others [32,33]. The production of a chairside dental model through a 3D printing method allows a quick reference to the dentist after the virtual designing is completed, thus, facilitating the treatment plan and communication between the dentist and patient [34]. A surgical guide for implant placement which is produced by the additive manufacturing method will allow the precise placement of a dental implant. This technique will help to eliminate a possible complication in which the vital nerve and blood vessels are traumatized. It will instead permit a “prosthesis-driven implant placement”. Moreover, a surgical guide is a specifically designed tool which utilizes multiple specialized softwares to adopt the virtual simulation process prior to the surgical appointment [35,36]. Due to the limitation of subtractive manufacturing such as the milling process, 3D printing is considered as the solution from a scientific and technical point of view. The overall schematic flow of the utilization of an additive manufacturing is simplified in Figure 2 with the example emphasized in implant surgical guide construction and fabrication protocol. Owing to the rapid development and thriving evolution of the implant rehabilitation utilizing additive manufacturing, the author briefly discusses the contemporary update on the field particularly on the surgical static and dynamic system.

Artificial Intelligence (AI) is the ability of a machine to perform human tasks. It revolves around the ability of a machine, around its own intelligence, to solve problems based on the learning of a specific set of data. The foundation of AI is to increase the ability of machines or its intelligence components to perform tasks with speed, low resources, accuracy, and others [54,55]. It will also eliminate human intervention such as potential human error, emotion, and bias, thus, making it a perfect solution for laborious work with increased risk of error. Other potential human symptoms such as fatigue, tiredness, and boredom after a continuous repetitive work are also eliminated [4]. It cannot be emphasized enough that AI requires advanced machine learning on huge datasets (“big data”) as it utilizes specific algorithm to perform the required works [56].

The application of AI in dentistry is huge with enormous potential [56]. In dental radiology, the application of AI by using cone beam computed tomography allows an automated detection of specific landmarks such as in lateral cephalometric view, dental panoramic radiography, and others. The progress is also encouraging through other techniques such as automated detection of caries, periodontal disease, periapical disease, and detection of possible oral disease such as cysts or tumor [57]. In restorative dentistry, AI has been developed to suggest and detect the presence of dental caries [57]. The conventional method of the detection of caries includes a combination of visual tactile examinations with special investigations such as dental bitewing radiograph. Nonetheless, the evolution in AI challenges the normal paradigm in caries detection with the utilization of state of the art, multiple varieties of neural networks including convoluted neural network, artificial neural networks, and additional derivatives of neural network architecture such as ResNet 18 and ResNext50 [58]. A systematic review by Pravos-Privado et al. showed that despite the advancement of AI systems in caries detection, various heterogeneity aspects such as parameters, multiple neural network systems, and outcome complicates the accuracy, diagnostics performance, specificity, and sensitivity of AI in caries detection [59]. Another study by Hung et al. proved that machine learning can be utilized for root caries prediction and prognostic value on a large mass of population data using numerous demographics, general and oral health, social, and lifestyle variables [60]. Another machine learning algorithm, support vector machine (SVM) showed high predictability and accuracy in assessing the level of complexity of root canal treatments based on American Association of Endodontics Case Difficulty Assessment Forms and thus, are able to aid the classification, clinical cognitive decision by the clinician, and the overall rapid referral process done by general dental practitioners [61]. In an endodontics diagnostics procedure, the advancement of AI was utilized to detect the periapical pathology by using a prototype of deep convolutional neural network through mass data from CBCT with a high reliability in comparison with the manual conventional segmentation method [62]. This is due to the ability of deep learning and convoluted neural network of AI components to segment and integrate the original data from radiographic CBCT views and images which then further coded these formats into a form that can be evaluated later on. The advanced development of computer aided design and computer aided manufacturing also allows software to precisely construct dental prostheses despite complex geometrical shapes and laborious laboratory workflow with the potential of a high risk of error involved [63]. The dental restoration must be a perfect fit, able to undergo ideal function, and also be aesthetically pleasing. In orthodontics-driven AI, a diagnosis, monitoring, and a specific yet individualized treatment plan is available. The clear aligners are produced based on accurate 3D model scanning and dental models. The AI creates an algorithm in which it can predict and decide future tooth movement, and the necessary pressure to be applied to the teeth with aid and input from the dental practitioner [64]. AI is designed to facilitate the construction of a surgical guide in specialized software with its major role in detecting the thickness, the height, and the density of the bone based on data acquisition from cone beam computed tomography [11]. This will facilitate the dental practitioner’s decision on suitable timing and technique for the implant placement. Diagnosis of temporomandibular joint disorders (TMDs) can be detected based primarily on history and signs and symptoms, which are then to be followed by clinical examinations. Theoretically, the collective clinical data of TMDs examinations, if they can be translated into a well-structured and organized computer language, are able to differentiate between absolute TMDs diagnosis and other clinical conditions mimicking TMDs [65]. A study by Shoukri et al. (2019) showed that the neural networks are able to program and classify the TMDs based on the combination of condylar radiographic imaging utilizing CBCT, biological markers such as saliva, and a variable range of clinical indicators including detailed facial and muscle pain and soreness history, range of mouth opening, and other signs such as headaches [66].

The application of AR/VR in healthcare services, particularly in dentistry, may give rise to issues such as enormous, massive data availability and trusted sharing. The privacy of a patient’s data is handled by a series of systems utilizing software type algorithms intended to represent human cognitive processes in clinical decision-making. The dental practitioner and the auxiliary teams are responsible for data handling with potential risks of data privacy and security breaches. The AI/VR systems are not held accountable for this, though the application is performed either under supervision or not within the legal and jurisprudence context [65,67].

The emerging wealth disparities among communities widens each year further representing the economic distribution and inequalities. Theoretically, the utilization of AR/VR and AI will streamline the workforce and reduce laborious, repetitive tasks and the need for additional manpower. In the long run, it is predominantly an effective way to reduce the operating cost and increase the revenues for healthcare industries. Nevertheless, the shift of physical, repetitive jobs to a more complex, cognitive driven one as required in the globalized society and industries will effectively reduce the need for human involvement in low-scale, laborious tasks causing depletion of available jobs in the healthcare sector. Hence, there will be issues in a fair, post-labor economy with division of an income-based society in the future [68].

The infrastructure support like computing power and requirements are critical to ensure the smooth processing of data updating, gathering, and interpretation in the delivery of the oral health care system which utilizes AI. The continuous expanding data of patients including demographic, clinical, treatment, and follow up requires consistent upgrading of the computing power which potentially may give rise to health, economic, training, technical logistics, and maintenance issues [65,69]. The inability to match the expected computational resources will reduce the efficacy of the AI, thus, reducing the delivery of the modern healthcare services. With that in mind, the common ideal solution for this is to utilize quantum supercomputing which can process the conventional binary bit using quantum version (qubits) which is fundamentally faster than conventional computing systems [70,71].

The form of population-derived health records linked at the personalized individual level of information and data is a great model for the health economic policy. The health data can be derived from conventional medical and dental screening, routine check-ups, follow-ups, and hospitalization together with other determinants such as socio-economic (income, jobs) and other social aspects (housing, food, security) [56]. Additionally, any other public information such as participation of the patient in online surveys, research, and forums in the context of Internet of medical things (IoMT) can be further evaluated and gathered [56]. This can be facilitated by the exponential usage of digital devices including smartphone, laptop, smartwatches, and digital online application facilitated by a superfast internet connection. Mass digital dental data and records are advantageous as they can be utilized by stakeholders for analytical disease prediction and prognostics modeling, population preventative programs, clinical research and surveys, clinical support systems, association of factors and cofounding factors for disease, identification of novel disease and treatment concepts, and the overall governance and delivery of the oral healthcare system. These population-based, yet personalized electronic digital oral health records will be an excellent platform and medium for interconnection between the dentist, dental auxillary team such as dental therapist and dental nurses, specialists, and physicians to understand, collaborate, and deliver an optimum level of oral healthcare through an interdisciplinary manner [56,72,73].

The conventional impression has played a part in the prosthodontics field in which the main aim of the procedure is to replicate and simulate the functional anatomical parts intraorally. Nevertheless, the challenge lies in producing it due to material disadvantages, namely volumetric shrinkage and expansion of dental stone and silicones, technique sensitivity and critical handling having to be done by the clinician, and others. This procedure is thus prone to error and inadvertently will affect the final prostheses outcome [74,75,76].

The current development of computer-aided design and computer-aided manufacturing (CAD-CAM) has brought the prosthodontics field to a new frontier. The construction of dental prostheses has been rapidly evolving with the influence of subtractive manufacturing with the example of machinable milling systems and additive manufacturing in 3D printers that have resulted in rapid prosthesis production and shortened manufacturing time. Three-dimensional (3D) dental models are obtained by intraoral scanners, thus, eliminating the need for the conventional impression method. The advantages include less time required for the impression taking procedure, avoiding cross-contamination risks, better communication tools for dental technicians and patients, as well as more simplified procedures for the clinician [76,77].

Dental prostheses are made available through direct fabrication using a CAD-CAM system after obtaining a digital impression with an intraoral scanner. This will eliminate the need for physical models. However, manual fabrication and the construction of dental prostheses are still required in the construction of lithium disilicate restoration in which manual veneering wax-up must be done and the casting of metal alloy for full or partial coverage for gold restoration [78]. An extraoral laboratory scanner then scans the dental impression or gypsum casts to create a 3D model, and the restoration is then designed on computer with a specialized design software and then 3D printed. This protocol is beneficial to the dental technician and even useful for clinicians who are still using the conventional impression method but require prostheses that adopt the CAD-CAM approach with the example of monolithic zirconia restoration [79].

It is paramount to evaluate the accuracy of these digital approaches, especially the intraoral scanner since it is the first step in adopting the CAD-CAM approach. Multiple studies had evaluated the accuracy of the intraoral scanner; however, they were limited to the production of orthodontic appliances such as retainers and implant prosthodontics [80,81,82,83]. As defined by International Organization for Standardization (ISO-5725), accuracy consists of trueness and precision. Trueness refers to the closeness of the experimental result to the true value. High trueness indicates that the experimental result is very close or equivalent to the true value. In contrast, precision describes the closeness of agreement between intragroup data obtained by repetitive measurements [84]. A number of studies had evaluated the accuracy between intraoral scanners for fixed dental prosthesis; nevertheless, the results are debatable. Zimmermann et al. reported that each intraoral scanner with a different CAD system has a different value of accuracy [85]. Kim et al. reported that CS3600 by Carestream Dental USA and conventional impression with laboratory scanning had the highest accuracy in production of zirconia crown; however, this study does not include other relevant, commercially available intraoral and extraoral scanners [86]. Nedelcu et al. showed that in the production of a distinctive finish line in full coverage crown scanning, CareStream 3600 and Trios 3 by 3 Shape showed better accuracy although this study focused on finishing line integrity [87]. Meanwhile, Planmeca Planscan intraoral scanning devices showed the least accurate digital impression by Mennito et al. [88]. Another elaborative study by Mangano et al., in full arch implant impression, comparing the trueness of 12 different intraoral scanners utilizing multiple variables outcomes including mesh/mesh evaluation and nurbs/nurbs evaluation showed that there was a significant difference in terms of accuracy between different intraoral scanners that are commercially available worldwide [89].

The delivery of modern oral healthcare should be derived based on modern technology driven by a patient-centered outcome. Digitalization in dentistry will facilitate oral healthcare to an optimum level. The pandemic of COVID-19 showed that tele-dentistry with remote consultation and artificial intelligence has a major role to play. It will indefinitely reduce the unnecessary contact between the patients and healthcare providers, shorten the duration of treatment, and be more cost effective in the long run. The field of dentistry is most likely to benefit especially in the utilization of AR/VR and AI systems for the delivery of pedagogy and clinical skills teaching. The research on digitalization in healthcare especially in dentistry should be the main focus in the next few decades with the aim of improving data acquisition and big datasets, safety and security of the “Big Data”, updating the neural networks, machined and deep learning of artificial intelligence, and other relevant fields.

 

This entry are from The Modern and Digital Transformation of Oral Health Care: A Mini Review