ABSTRACT
The application of artificial intelligence (AI), which includes its limitations and the expected development of dental diagnostics based on it, as well as more efficient treatment planning and image analysis, prediction of possible treatment outcomes, and record-keeping, is the main topic of this chapter. As more and more AI-based applications focus on patient care and relieving dentists of strenuous routine tasks, they help improve patient health, reduce treatment costs, and enable the intensive development of personalized, predictive, preventive, and participatory dentistry. Although this technology belongs to the future, today its wide application is still rare, due to limited data accessibility, and lack of methodology and necessary standards for their application. Therefore, this chapter should serve as an initial capsule that helps this technology become a very serious topic to bridge the gap between its possibilities and its everyday application as soon as possible. This chapter also describes how, with the help of AI, to automate the assessment of a radiographic image in order to detect a dental disease faster and more accurately. It has been shown how appropriately taught neural networks (NN) can be an extremely successful aid for the diagnostician and how AI could shape the future of public health and care delivery. AI, especially deep learning, is based on skillful machines to imitate our cognitive functions in performing various functions by using convenient software, which mimics the human brain. This software can learn from data to make the most successful assessments of disease development and successfully diagnose diseases, based on irregularity in radiographic images. It is also shown how these systems save the workload of radiologists, speed up the recording and presentation of data, give useful answers regarding the best treatment methods, and reduce the risk of cognitive biases. Various specific applications of such methods in different dentinal and maxillofacial radiology branches have been shown, as they enable the correct interpretation of very complex images and the location of changes in the corresponding tissues.
