Artificial intelligence and other modern digital technologies for detection of dental fluorosis: A review of literature

ABSTRACT Dental fluorosis has been one of the common dental hard tissue problems in humans. It is considered as public health concern because it affects aesthetics, form, and functions of affected teeth. Dental fluorosis is commonly detected or diagnosed by visual examination along with positive history of chronic high fluoride exposure during childhood. The literature emphasizes that the rapid and accurate diagnosis and classification of dental fluorosis is of paramount importance in clinical practice and epidemiological studies. There have been various traditional methods or indexes for the diagnosis of the dental fluorosis. However, in today’s modern era, digital technological advancement is rapidly evolving in health sector including dentistry. Contemporary dentistry is moving toward modern digital tools/methods for detection of oro-dental problems and appropriate treatment. Recently, novel digital methods for detection and diagnosis of dental fluorosis have been described and reported by the studies. However, there is lack of review article regarding applications of modern digital technologies in the evaluation of dental fluorosis. Hence, the present review article aimed to explore the relevant literature pertaining to newer digital technologies for detection/diagnosis or evaluation of dental fluorosis condition and then summarize in a descriptive (narrative) manner. In this review article, various aspects and prospects about “artificial intelligence (AI),” “Machine learning (ML),” “Deep Learning (DL),” and some other novel emerging techniques (such as, “Fuzzy k-nearest neighbor method/technique,” “Fourier-transform infrared spectroscopy (FTIR),” “Cuckoo Search algorithm,” “PCA-LDA,” “Raman micro-spectroscopy,” and “Bayesian hurdle mixed-effects model”) have been discussed particularly in context of detection or diagnosis of dental fluorosis.